Documente Academic
Documente Profesional
Documente Cultură
Chapter 4
Arterial, Central Venous, and
4
53
Arterial
Pressure
B C
Figure 4-1. The systolic pressure variation (SPV) is the concern about infection.4 Use of other sites, such as the
difference between the maximum and the minimum systolic brachial or the axillary artery or even the dorsalis pedis
blood pressures during one ventilatory cycle (A). The SPV is artery,5 can be considered. An important point to keep in
the sum of the ∆Down (B) and the ∆Up (C). (From Perel A,
mind is that the pulse pressure increases from the core to
Pizov R, Cotev S: Systolic blood pressure variation is a
sensitive indicator of hypovolemia in ventilated dogs the periphery. In other words, the systolic pressure is over-
subjected to graded hemorrhage. Anesthesiology estimated in smaller arteries (Fig. 4-2). Hence, it may be
1987;67:498.) better to rely more on mean than on systolic or diastolic
pressures.
54
55
ECG
Pressure, mm Hg
30 2
3
20
1 a v
4 x y
10
ac v 6
x y
0 5
1 sec
Figure 4-3. Pressure waveforms. A, The normal right atrial (RA) tracing. The a wave is the RA pressure rise resulting from atrial
contraction and follows the P wave of the electrocardiogram (ECG). On simultaneous ECG and RA tracings it usually occurs at
the beginning of the QRS complex. The c wave, caused by closure of the tricuspid valve, follows the a wave and is coincident
with the beginning of ventricular systole. Atrial relaxation (x descent) is followed by a passive rise in RA pressure resulting from
atrial filling during ventricular systole and occurs during the T wave of the simultaneously recorded ECG. The y descent reflects
the opening of the tricuspid valve and passive atrial emptying. B, The normal right ventricular (RV) tracing. The sharp rise in RV
pressure (1) is due to isometric contraction and is followed by a rapid pressure decrease (2) as blood is ejected through the
pulmonary valve. This rapid ejection is followed by a phase of more reduced pressure decrease, which is often reflected in
a small step in the downslope of the RV pressure waveform (3). The subsequent sharp decline in RV pressure (4) occurs as
a result of isometric relaxation, and is noted once the RV pressure falls below the pulmonary artery (PA) pressure (with
consequent closure of the pulmonary valve). As RV pressure falls below RA pressure, the tricuspid valve opens, and passive
refilling (5) of the right ventricle occurs, followed by atrial contraction, causing a biphasic wave of ventricular filling to appear
on the RV tracing (6). C, The normal pulmonary arterial waveform. A pulmonary artery systolic elevation is caused by ejection
of blood from the right ventricle, followed by a decline in pressure as RV pressure falls. As RV pressure falls below pulmonary
artery pressure, the pulmonary valve closes, which causes a momentary rise in the declining pulmonary artery pressure. This is
the dicrotic notch characteristic of the pulmonary arterial (and also the systemic arterial) waveform. The pulmonary artery
systolic wave usually occurs in synchrony with the T wave of the ECG. Pulmonary artery diastolic pressure (PADP) does not fall
below RA pressure and therefore is higher than right ventricular end diastolic pressure (RVEDP): it is an approximation to left
ventricular end diastolic pressure (LVEDP). D, The normal pulmonary artery occlusion pressure (PAOP) waveform. The waveform
of the PCWP is subject to the same mechanical variables as the RA waveform, but because of the damping that occurs through
the pulmonary circulation, the waves and descents often are less distinct. Similarly, the mechanical events are recorded later in
the cardiac cycle, as seen on the ECG. Thus, the a wave is not seen until after the QRS complex, and the v wave occurs after
the T wave of the ECG. The PCWP is a closer approximation to LVEDP than is PADP. (From Grossman W: Cardiac
catheterization. In Braunwald E [ed]: Heart Disease: A Textbook of Cardiovascular Medicine, 3rd ed. Philadelphia, Saunders,
1992.)
pressure within the pulmonary veins, these changes should limitations. Simple disconnection of the respirator to
be identical, because the pulmonary artery and vein measure PAOP after obtaining an equilibration state is
should be subjected to identical changes in intrathoracic not recommended, because the measured PAOP will not
pressures. If, on the other hand, the catheter tip is not in represent the real value when PEEP is applied. A better
a West zone III, the changes in PAOP will be more signifi- method consists of measuring the lowest (nadir) PAOP
cant than the changes in PAP. In these latter conditions, within seconds after a very transient disconnection from
either fluid administration or some reduction in the PEEP the ventilator, to identify the true transmural pressure
level may abolish the differences. before a new equilibrium is reached.14 Such a maneuver
The next question is whether the measured pressure is suppresses the high intrathoracic pressure and eliminates
truly a transmural pressure—that is, the pressure differ- the influence of the extramural pressure, but the values
ence between the vascular structures and the environ- obtained may not be valid in the presence of intrinsic
mental structures. In other words, will changes in PEEP. Other methods have been suggested, some rela-
surrounding intrathoracic pressures influence the pres- tively sophisticated15 and others more simple, involving
sures measured? In the case of intrathoracic pressures, the subtraction of approximately one third of the PEEP level
changes in pleural pressure are particularly relevant. from the measured PAOP, for example. The question is
Hence, all measurements should be performed at end- whether this is really so important, because absolute
expiration, when the pleural pressure is closest to zero. A PAOP values are not very helpful.
marked fall in intrapleural pressure, as with a Mueller In respiratory failure, PAOP does not represent true
maneuver (forced inspiration against resistance), may dra- capillary pressure, which may be somewhat higher. The
matically increase the transmural pressures. A more true capillary pressure may be estimated from the mea-
common problem in the intensive care unit (ICU) is the surement of the intersection point of the rapid and slow
increase in pleural pressure due to positive-pressure ven- pressure decay curves recorded after a rapid interruption
tilation, sometimes with high PEEP levels. of the blood flow; such measurements of capillary pres-
One method of combating this problem could be to sure are possible from the pressure trace. This pressure is
subtract the esophageal pressure from the measured well correlated with extravascular lung water in animal
PAOP, but this approach has a number of technologic experiments. The need to know these values is question-
56
Cardiac Output
The reference method for measurement of cardiac output
is use of the Fick equation; however, this is difficult to
Figure 4-4. The position of the pulmonary artery catheter tip apply in practice. The indicator dilution technique has
in relation to the zones of the theoretical lung model been used instead. Indocyanine green clearance has been
described by West. For pulmonary artery occlusion pressure used for many years, but this method is time-consuming
(PAOP) to be a valid estimate of “left heart” filling pressure, a and quite difficult to perform. The thermodilution tech-
continuous column of blood must exist between the catheter
nique developed by Ganz is a convenient technique
tip and the left atrium. In zones I and II, pulmonary
vasculature is either totally or partially compressed by the which today allows the almost continuous measurement
intra-alveolar pressure, and measurements of PAOP will be of cardiac output. The presence of tricuspid insufficiency
misleading. PA, alveolar pressure; Pa, pulmonary artery is the major limitation to this technique.
pressure; Pv, pulmonary venous pressure. (From Marini JJ: More recently, other techniques, including the transpul-
Respiratory Medicine and Intensive Care for the House Officer. monary technique proposed by PiCCO (Munich, Germany)
Baltimore, Williams & Wilkins, 1991.) and the lithium dilution technique proposed by LiDCO
(London), have been developed, but they are somewhat
able, however, because the primary goal remains to keep less precise.
these hydrostatic pressures as low as possible while main- The thermodilution technique estimates cardiac output
taining adequate cardiac output. over several cardiac cycles, whereas other techniques
PAOP may not adequately reflect LV end diastolic pres- including those based on pulse contour analysis, may
sure. It may be lower in patients with aortic regurgitation assess beat-to-beat variations. These may be useful to
or higher in the presence of significant tachycardia or estimate the influence of changes in intrathoracic pres-
mitral valve disease. LV end diastolic pressure may not sures on the stroke volume variation, an estimate of fluid
even accurately predict LV end diastolic volume. This was responsiveness.
already demonstrated many years ago with radionuclide
techniques.16 Preload is more directly defined as the end Right Ventricular Volumes
diastolic volume. A stiff, noncompliant ventricle can result The use of a modified PAC equipped with a fast response
in a relatively high end diastolic pressure for a given end thermistor also allows evaluation of the right ventricular
diastolic volume. An evaluation of end diastolic volumes ejection fraction (RVEF) (Fig. 4-5). With knowledge of the
can be obtained less invasively with echocardiographic stroke volume, it becomes easy to calculate the end sys-
techniques. Likewise, the use of transthoracic thermodilu- tolic and end diastolic volumes. This measurement can be
tion techniques allows the estimation of intrathoracic particularly useful in the presence of RV failure, but this
blood volume and global end diastolic volumes. However, also is a situation in which the measurement is least reli-
these assessments of end diastolic volumes do not give able: tricuspid regurgitation secondary to pulmonary
additional information about the likelihood of fluid hypertension.
responsiveness.
In sum, then, a given level of cardiac filling pressures Mixed Venous Oxygen Saturation
does not provide much information about fluid respon- SvO2 represents the balance between oxygen consump-
siveness, but monitoring can be very helpful to guide a tion and oxygen supply. According to the Fick equation:
fluid challenge.17 During fluid administration, the goal is
VO2 = cardiac output × (CaO2 − CvO2)
to obtain a significant increase in cardiac output (by
the Frank-Starling mechanism) with the least increment where VO2 is oxygen uptake and CaO2 and CvO2 are the
in cardiac filling pressures, in order to minimize the arterial and venous oxygen content, respectively. If the
risk of edema formation. The goal is not to keep the dissolved oxygen in the blood is neglected for the pur-
cardiac filling pressures within predefined arbitrary limits; poses of calculation, then
57
EF Dysfunction
Failure
Closing
End-Diastolic Volume Pressure
Figure 4-5. Ventricular function curve depicting the difference
between ventricular dysfunction (decrease in the ventricular
ejection fraction [EF] but preservation of stroke volume) and
Flow
ventricular failure (defined as the inability of the heart to
pump enough flow). The EF, the ratio between stroke volume Figure 4-6. Relationship between pressure and blood flow,
and end diastolic volume, is represented by the dashed lines; illustrating the limitations in the calculation of the vascular
it is decreased in all cases. resistance. Note that the line defining that relationship does
not go through the origin, since the pressure is not zero in
the absence of blood flow.
58
4 Box 4-1
B
59
Balloon perforation
Small branch
range of frequency values largely exceeding the range of connectors, may decrease the resonant frequency,
frequencies found in the human body. resulting in “whipped” traces. Likewise, the presence
Reliability is less secure in the intermediary system of bubbles must be carefully avoided. Transient flushing
made of tubes and stopcocks along the extent of the pres- should be followed by an abrupt return of the pressure
sure system. These can modify the morphology of the trace to its actual value.
trace, resulting in damping. Motion artifacts can further
complicate th e tracings. The presence of air bubbles also Applications: Diagnosis versus Monitoring
may alter the signal. Use of fluid-filled catheters to Today the PAC is more useful in guiding therapy rather
measure pressures provides reliable estimates of mean than in identifying abnormalities, this latter role having
vascular pressures. largely been taken over by noninvasive, mainly echocar-
Three steps must be followed to guarantee reliable diographic techniques (Table 4-1). In the past, analysis of
measurements: waveforms was used—for example, the abnormal V waves
of mitral regurgitation. Likewise, an increase of all pres-
1. The first step is appropriate zeroing. This is accom-
sures to identical levels should suggest pericardial tam-
plished by opening the transducer to atmospheric pres-
ponade. These findings should still alert the clinician to
sure (taken as the zero value). All pressures must be
possible abnormalities, but echocardiographic techniques
measured with reference to an arbitrary reference
have largely replaced the use of the PAC for identifying
point. This zero reference pressure level should ideally
valvular disease. The use of echocardiographic techniques
be where it is least influenced by location on the body.
for monitoring, however, is hampered by the difficulty of
In humans, it is thought to be at the level of the right
keeping the probe in the esophagus for prolonged periods
atrium, so the reference level usually is placed in the
of time, and results are very operator dependent.
mid-chest (mid-axillary) position at the level of the
To illustrate the need to integrate different variables,
fourth intercostal space. In healthy persons, the CVP
rather than focusing on just one variable, some suggested
referred to that region does not change with supine
applications follow:
versus upright position. An alternative reference is
5 cm vertically below the sternal angle. Obviously, ■ Interpretation of a cardiac output value: It is impor-
errors in zeroing are relatively more important for tant to consider the four determinants of cardiac output
measurements of cardiac filling pressures than for arte- (Fig. 4-9). Interpretation should start by considering
rial pressure measurements, because the errors are the stroke volume (cardiac output divided by heart
quantitatively identical but proportionally greater. rate), and relating this to an index of ventricular filling
2. The second step is calibration, which is now done auto- (PAOP) and an index of ventricular afterload (arterial
matically by today’s electronic systems. and pulmonary artery pressures); an inotropic agent
3. The third step is ensuring the good quality of the trace. should be added only when preload and afterload have
Shaking the catheter should result in large pressure been optimized.
variations on the screen. A damped system will under- ■ Fluid status: Low cardiac filling pressures may be
estimate systolic pressures. The liquid column should normal or reflect hypovolemia. It is the measurement
be continuous, without air bubbles in the system. of cardiac output and SvO2 that will help determine
Excessive tubing length, or multiple stopcocks and fluid needs. Indeed, hypovolemia typically is
60
PA Catheterization Echocardiography
Disorder Diagnosis Monitoring Diagnosis Monitoring
Tamponade + ++ +++ +
Hypovolemia ++ +++ ++ +
Valvular disease + ++ +++ +
Heart failure ++ +++ ++ +
Right ventricular failure ++ +++ +++ ++
Septic shock + +++ + +
Increased O2 extraction
Contractility
Afterload CO high CO low
Figure 4-9. The four determinants of cardiac output.
Hemoglobin PAOP high PAOP low
associated with a low cardiac output and a low SvO2
(Fig. 4-10). By contrast, hypervolemia in the presence
Normal Low Heart failure Hypovolemia
of normal cardiac function will be manifested by high (anemia)
cardiac filling pressures associated with a relatively
high cardiac output and normal or high SvO2. Elevated VO2
■ Hemodynamic versus nonhemodynamic pulmonary (exercise, stress, anxiety)
edema: The distinction between hemodynamic and Figure 4-10. Interpretation of a low SvO2 (mixed venous
nonhemodynamic types of lung edema no longer oxygen saturation). CO, cardiac output; PAOP, pulmonary
requires pulmonary artery catheterization; the clinical artery occlusion pressure; SaO2, arterial oxygen saturation;
history and less invasive (e.g., echocardiographic) mea- VO2, oxygen uptake.
surements usually are sufficient to separate the two.
Nevertheless, invasive hemodynamic monitoring can
RAP > PAOP
reveal that patients thought to meet only the acute
lung injury/acute respiratory distress syndrome (ALI/
ARDS) criteria sometimes have unexpectedly high Cardiac Output
PAOP.20 Adequate Inadequate
■ RV dysfunction versus failure: A reverse gradient (adequate SvO2) (low SvO2)
between RAP and PAOP (i.e., a higher RAP than
PAOP) indicates RV dysfunction or failure and usually RV Dysfunction RV Failure
is secondary to pulmonary hypertension, as will be
immediately apparent from the measurements of pul- Pulm. Art. P.
+ Inhaled NO?
monary artery pressures. RV dysfunction is manifested PGE1?
Normal Increased
by RV dilation (and thus a decrease in RVEF) with no
limitation on cardiac output. This is the most common Nothing
Diuretics? Arterial Pressure
situation in patients with ARDS, who usually maintain
a hyper-kinetic state. Rather, RV failure refers to a state Low Normal
in which cardiac output is no longer sustained at ade- Dobutamine
quate levels.21 These different entities are illustrated in Vasopressors Vasodilators
Figure 4-11.
Figure 4-11. Interpretation of a reversed gradient between
■ LV dysfunction versus failure: Here the gradient
the right atrial pressure (RAP) and the pulmonary artery
between PAOP and RAP will be higher than the typical occlusion pressure (PAOP). NO, nitric oxide; PEG1,
3 to 5 mm Hg. As with the right ventricle, in LV dys- prostaglandin E1; RV, right ventricular; SvO2, mixed venous
function the ventricular volumes may increase (so the oxygen saturation.
61
exist in septic shock. LV failure is more common and is is difficult to wean from mechanical ventilation, possi-
manifested by a decrease in cardiac output and SvO2. bly owing in part to cardiac dysfunction.
62
60 60
40 40
20 20
0 0
Figure 4-13. Partial pulmonary artery occlusion pressure (Ppao) measured when 1.5 cc of air was used to inflate the catheter
balloon (left graph); a much lower Ppao was obtained with a 1.0-cc inflation (right graph). Review of the chest roentgenogram
revealed that the catheter tip was too peripheral, so it was withdrawn to a more proximal location. Ppa, pulmonary artery
pressure. See text for definition of partial and best Ppao values. Scale in mm Hg. (From Leatherman JW, Shapiro RS:
Overestimation of pulmonary artery occlusion pressure in pulmonary hypertension due to partial occlusion. Crit Care Med
2003;31:93-97.)
63
1. Bur A, Hirschl MM, Herkner H, et al: impregnated central venous catheters. 22. Rhodes A, Lamb FJ, Malagon I, et al:
Accuracy of oscillometric blood pressure Catheter Study Group. N Engl J Med A prospective study of the use of a
measurement according to the relation 1999;340:1-8. dobutamine stress test to identify
between cuff size and upper-arm 12. Tambe SM, Sampath L, Modak SM: outcome in patients with sepsis, severe
circumference in critically ill patients. In vitro evaluation of the risk sepsis, or septic shock. Crit Care Med
Crit Care Med 2000;28:371-376. of developing bacterial resistance 1999;27:2361-2366.
2. De Backer D, Heenen S, Piagnerelli M, to antiseptics and antibiotics 23. Sakr Y, Vincent JL, Reinhart K, et al: Use
et al: Pulse pressure variations to predict used in medical devices. J of the pulmonary artery catheter is not
fluid responsiveness: Influence of tidal Antimicrob Chemother 2001;47: associated with worse outcome in the
volume. Intensive Care Med 589-598. intensive care unit. Chest 2005;128:
2005;31:517-523. 13. O’Grady NP, Alexander M, Dellinger EP, 2722-2731.
3. Heenen S, De Backer D, Vincent JL: et al: Guidelines for the prevention of 24. Harvey S, Harrison DA, Singer M, et al:
How can the response to volume intravascular catheter–related infections. Assessment of the clinical effectiveness
expansion in patients with spontaneous Infect Control Hosp Epidemiol of pulmonary artery catheters in
respiratory movements be predicted? 2002;23:759-769. management of patients in intensive
Crit Care 2006;10:R102. 14. Pinsky MR, Vincent JL, De Smet JM: care (PAC-Man): A randomised
4. Lorente L, Santacreu R, Martin MM, et al: Estimating left ventricular filling controlled trial. Lancet 2005;366:
Arterial catheter-related infection of pressure during positive end-expiratory 472-477.
2,949 catheters. Crit Care 2006;10:R83. pressure in humans. Am Rev Respir Dis 25. Rhodes A, Cusack RJ, Newman PJ, et al:
5. Martin C, Saux P, Papazian L, et al: 1991;143:25-31. A randomised, controlled trial of the
Long-term arterial cannulation in ICU 15. Teboul JL, Pinsky MR, Mercat A, et al: pulmonary artery catheter in critically ill
patients using the radial artery or Estimating cardiac filling pressure in patients. Intensive Care Med
dorsalis pedis artery. Chest 2001;119: mechanically ventilated patients with 2002;28:256-264.
901-906. hyperinflation. Crit Care Med 26. Mimoz O, Rauss A, Rekik N, et al:
6. Cook DJ, Simel DL: The rational clinical 2000;28:3631-3636. Pulmonary artery catheterization in
examination. Does this patient have 16. Calvin JE, Driedger AA, Sibbald WJ: critically ill patients: A prospective
abnormal central venous pressure? Does the pulmonary capillary wedge analysis of outcome changes associated
JAMA 1996;275:630-634. pressure predict left ventricular preload with catheter-prompted changes in
7. Rivers E, Nguyen B, Havstad S, et al: in critically ill patients? Crit Care Med therapy. Crit Care Med 1994;22:
Early goal-directed therapy in the 1981;9:437-443. 573-579.
treatment of severe sepsis and septic 17. Vincent JL, Weil MH: Fluid challenge 27. Chittock DR, Dhingra VK, Ronco JJ,
shock. N Engl J Med 2001;345: revisited. Crit Care Med 2006;34: et al: Severity of illness and risk of death
1368-1377. 1333-1337. associated with pulmonary artery
8. Joynt GM, Gomersall CD, Buckley TA, 18. Varpula M, Karlsson S, Ruokonen E, catheter use. Crit Care Med 2004;32:
et al: Comparison of intrathoracic and et al: Mixed venous oxygen saturation 911-915.
intra-abdominal measurements of cannot be estimated by central venous 28. Friese RS, Shafi S, Gentilello LM:
central venous pressure. Lancet oxygen saturation in septic shock. Pulmonary artery catheter use is
1996;347:1155-1157. Intensive Care Med 2006;32: associated with reduced mortality in
9. Merrer J, De Jonghe B, Golliot F, et al: 1336-1343. severely injured patients: A National
Complications of femoral and 19. Yalavatti GS, De Backer D, Vincent JL: Trauma Data Bank analysis of 53,312
subclavian venous catheterization in The assessment of cardiac index in patients. Crit Care Med 2006;34:
critically ill patients: A randomized anemic patients. Chest 2000;118: 1597-1601.
controlled trial. JAMA 2001;286: 782-787. 29. Morris AH, Chapman RH, Gardner RM:
700-707. 20. Wheeler AP, Bernard GR, Thompson BT, Frequency of wedge pressure errors
10. Zurcher M, Tramer MR, Walder B: et al: Pulmonary-artery versus central in the ICU. Crit Care Med 1985;13:
Colonization and bloodstream infection venous catheter to guide treatment of 705-708.
with single- versus multi-lumen central acute lung injury. N Engl J Med 30. Al Kharrat T, Zarich S, Amoateng-
venous catheters: A quantitative 2006;354:2213-2224. Adjepong Y, et al: Analysis of observer
systematic review. Anesth Analg 21. Vincent JL: Is ARDS usually associated variability in measurement of
2004;99:177-182. with right ventricular dysfunction or pulmonary artery occlusion pressures.
11. Darouiche RO, Raad II, Heard SO, et al: failure? Intensive Care Med Am J Respir Crit Care Med 1999;160:
A comparison of two antimicrobial- 1995;21:195-196. 415-420.
64