Bongard Frederic , Darryl Sue. Diagnosis and Treatment Critical Care

Подождите немного. Документ загружается.



187

008

Intensive Care Monitoring

Kenneth Waxman, MD

Frederic S. Bongard, MD

Darryl Y. Sue, MD

Physiologic monitoring is available for appropriate indica-

tions in the ICU. Monitoring should be selected and applied

to detect pathophysiologic abnormalities in patients at high

risk of developing them and to aid in the titration of therapy

to appropriate physiologic end points.



Electrocardiography

Continuous electrocardiography permits monitoring of heart

rate, detection of arrhythmias, and evaluation of pacemaker

function. It also may help detect myocardial ischemia or elec-

trolyte abnormalities. Continuous electrocardiographic mon-

itoring is indicated for patients with potential for developing

arrhythmias—particularly those with acute myocardial

infarction, traumatic cardiac contusion, following cardiac

surgical procedures, and those with a prior history of arrhyth-

mia. It is also useful for those in whom heart rate monitoring

is indicated, such as patients at risk of hemorrhage or those

undergoing fluid resuscitation. Monitoring of the ST seg-

ments is indicated for patients at risk of myocardial ischemia,

such as those with coronary artery disease who have an injury,

illness, or operation. Monitoring of the ECG also may be use-

ful to detect certain electrolyte abnormalities such as

hypokalemia during treatment of diabetic ketoacidosis.

The cardiac electrical potential available for skin surface

monitoring is between 0.5 and 2.0 mV. Because of this low

signal level, electrocardiographic systems must have good

sensing, amplifying, and display capabilities. The electrodes

used for electrocardiographic monitoring are usually com-

posed of silver–silver chloride gel (Ag/AgCl) inside an adhe-

sive pad. Prior to placement, the skin should be clean and

dry. The stratum granulosum has an electrical resistance of

50,000 Ω/cm

, which can be reduced to 10,000 Ω/cm

simply

by cleansing, which removes oils and dead cells. Difficulties

with a low signal are often remedied by reapplying the elec-

trode after cleaning the skin.

Optimal electrode placement (Figure 8–1) allows proper

detection of the electrocardiographic signals with a minimum

of extraneous noise. A “modified lead II” configuration is

appropriate for routine monitoring, with limb leads

extended proximally to lie over the shoulders. Placing them

over bony prominences reduces electrical noise from muscle

contractions.

Most electrocardiographic amplifiers and display modules

can be used for both diagnostic and monitoring applications.

The diagnostic setting permits greater amplifier bandwidth

(0.05–100 Hz) when compared with the monitor setting

(0.5–50 Hz). For routine rate and arrhythmia detection, the

monitor setting is preferred because it decreases baseline

wander, reduces unwanted interference, and improves overall

trace quality. However, because it may falsely elevate or

depress ST segments, the diagnostic mode should be selected

when myocardial ischemia is the primary concern.

Clinical Applications

A. Electrocardiographic Monitoring—Lead placement at

the shoulders and in the lead II position parallels the atria

and results in the greatest P-wave voltage of any surface lead

configuration. This facilitates recognition of arrhythmias

and inferior wall ischemia. When placed in the V

position

along the anterior axillary line, both anterior and lateral wall

ischemia can be detected. Because patient positioning may

make a true lead V

position difficult, a modified arrange-

ment (CS

), in which the left arm lead is placed just lateral to

the left nipple and the lower limb lead is placed over the iliac

crest, is a good alternative. When possible, leads II and V

should be monitored simultaneously. Esophageal leads are

better than lead II for the detection of arrhythmias, but their

use is difficult in patients who are not paralyzed and sedated,

and they are rarely used in the ICU setting.

B. Complications—Difficulties associated with electrocardio-

graphy are usually due to technical error or equipment mal-

function. Electrodes may not function properly when they are

old and dry or if they are not attached securely. Electrical noise



CHAPTER 8

188

accompanying the displayed electrocardiogram is usually due

to loose electrodes, broken wires, or poorly fitting contacts or

problems with associated electrical equipment. Proximity of

the patient to electrical lines (ie, power cords, etc.) produces a

potential difference through capacitive coupling known as

common-mode voltage. Typically as low as a few millivolts, cou-

pling can cause voltages as high as 20 V. Common-mode volt-

age interference usually presents as 60-Hz interference and

often can be reduced by using properly placed shielded wires,

good skin preparation, and an electrocardiographic amplifier

that offers common-mode rejection.

Proper sensitivity setting of the amplifier and recorder

is essential to make certain that large T waves are not

“double-counted” in the rate determination. Additional fil-

tering is occasionally required for patients who have

pacemakers—in whom the pacer spike is interpreted as a

QRS complex.

Goodman S, Shirov T, Weissman C: Supraventricular arrhythmias

in intensive care unit patients: Short and long-term conse-

quences. Anesth Analg 2007;104:880–6. [PMID: 17377100]

Reinelt P et al: Incidence and type of cardiac arrhythmias in criti-

cally ill patients: A single-center experience in a medical-

cardiological ICU. Intensive Care Med 2001;27:1466–73.

[PMID: 11685339]



Blood Pressure Monitoring

Because systemic blood pressure is related to both cardiac

function and the peripheral circulation, blood pressure mon-

itoring provides information related to the overall circula-

tory condition. While blood pressure monitoring is standard

and universal for critically ill patients, the type of blood pres-

sure monitoring and its frequency should be chosen based

on the individual patient’s diagnosis and condition.



Figure 8–1. Locations of the unipolar precordial leads on the body surface. (Reproduced, with permission, from

Goldschlager N, Goldman MJ: Principles of Clinical Electrocardiography, 13th ed. Originally published by Appleton

Midclavicular line

Anterior axillary line

Midaxillary line

Horizontal plane

of V

4-6

)(V

)



INTENSIVE CARE MONITORING

189

Blood pressure represents lateral force exerted on the vas-

culature by flowing blood. Pressure is maximal shortly after

ventricular systole (SBP). The diastolic pressure (DBP) fol-

lows cardiac diastole and is the lowest pressure in the cycle.

The mean arterial pressure (MAP) represents the standing

pressure in the arterial circuit and is calculated as follows:

Pulse pressure is the arithmetic difference between the sys-

tolic and diastolic pressures. Pulse pressures vary with stroke

volume or vascular compliance. Pulse pressures less than

30 mm Hg are common with hypovolemia, tachycardia, aor-

tic stenosis, constrictive pericarditis, pleural effusions, and

ascites. Widened pulse pressures may be due to aortic regur-

gitation, thyrotoxicosis, patent ductus arteriosus, arteriove-

nous fistula, and coarctation of the aorta. Variability of pulse

pressure and systolic pressure during the respiratory cycle has

been correlated with response to intravascular fluid repletion.

The initial upstroke and peak of the arterial waveform are

produced by left ventricular ejection. The end of systole is

marked by a brief decline in pressure until the aortic valve

closes and redirects backflowing blood into the aorta. The

“dicrotic notch” so created may be detected on recordings

obtained from aortic or proximal arterial sites. The waveform

becomes more peaked and of higher amplitude as it progresses

distally. The initial upstroke is prolonged, producing a higher

systolic and a lower diastolic pressure (Figure 8–2).

The velocity of blood flow is slowest in the largest arter-

ies because they are distensible and absorb energy from the

pressure wave front. The pulse wave travels at a rate of 7–10 m/s

in large arteries such as the subclavian artery and increases to

15–30 m/s in smaller distal arteries.

When a pressure wave front enters a small, nondistensible

artery, part of the wave may be reflected back proximally. If a

reflected wave strikes an oncoming wave, the two summate,

causing a higher pressure than would occur otherwise. This

phenomenon produces pressures in the distal peripheral

arteries that paradoxically may be more than 20–30 mm Hg

above those recorded in the aorta.

Arterial pressure depends on cardiac output (CO) and

systemic vascular resistance (SVR). The latter is calculated as

follows:

When MAP and CVP (central venous pressure) are in

millimeters of mercury and CO is in liters per minute, SVR

is expressed in dynes × seconds × cm

–5

. Evaluation of the

equation indicates that an increase in either SVR or CO will

increase mean arterial pressure.

Clinical Applications

Arterial blood pressure can be assessed either by direct

instrumentation of the vascular tree or by indirect means.

The indirect technique usually involves inflating a cuff to

occlude an artery. As the cuff is deflated and inflow resumes,

arterial pressure can be determined.

A. Noninvasive Arterial Pressure Monitoring—

1. Palpation—A blood pressure cuff is placed above an eas-

ily palpated artery and inflated until pulsation ceases. On

cuff deflation, systolic pressure is estimated as that pressure

at which pulsation resumes. This method is limited because

it underestimates true arterial pressure and does not provide

a diastolic pressure.

2. Auscultation (Riva-Rocci method)—When an occlud-

ing proximal cuff is deflated below systolic pressure, flow

begins through the compressed artery. The turbulent flow thus

created strikes the walls of the vessel, causing them to reverber-

ate (Korotkoff sounds). As long as the cuff pressure is higher

than diastolic pressure, no flow will occur during diastole. The

sound thus produced is rhythmic in nature. Once the pressure

in the cuff is below diastolic pressure, flow occurs throughout

the cardiac cycle, and the sounds disappear. A cuff 20% wider

than the diameter of the limb must be used to obtain correct

sphygmomanometric pressures. If the cuff is too narrow, the

systolic and diastolic pressures will be artificially increased,

and vice versa. Other sources of error include too-tight or too-

loose cuff application and too-rapid or too-slow cuff deflation.

SVR

(MAP CVP)

−

× 80

MAP

(SBP DBP)

+×2

Dorsalis pedis artery

Femoral artery

Radial artery

Peripheral

Central

Aortic root

Subclavian artery

Axillary artery

Brachial artery



Figure 8–2. The shape of the arterial pressure wave

front changes as it progresses distally. The systolic peak

becomes more pronounced, as does the dicrotic notch.

(Reproduced, with permission, from Morgan GE, Mikhail

MS: Clinical Anesthesiology. Originally published by

Companies, Inc.)



CHAPTER 8

190

Unduly slow deflation produces venous congestion, which

decreases the amplitude of the Korotkoff sounds as the cuff

pressure nears the diastolic pressure.

When compared with intraarterial pressure measure-

ments, those obtained by auscultation differ by 1–8 mm Hg

systolic and 8–10 mm Hg diastolic. At intraarterial pressures

below 120 mm Hg systolic, auscultation tends to overesti-

mate pressure, whereas above 120 mm Hg, auscultation

underestimates arterial pressure.

3. Oscillometry—The oscillometer uses two cuffs in series;

one occludes the artery proximally, whereas the other detects

the onset of pulsations. Slow deflation of the proximal cuff

produces aneroid needle oscillation or mercury column vari-

ation at systolic pressure. Oscillometry is the only noninva-

sive technique capable of indicating mean arterial pressure,

which coincides with maximum deflection of the manome-

ter. Although diastolic pressure is defined as that point at

which oscillation ceases, measurement of diastolic pressure is

in fact inaccurate. Oscillometry requires several cardiac

cycles to measure blood pressure accurately.

Automated oscillometry devices generally use single-

bladder cuffs that are alternately inflated and deflated. On

deflation, alterations in cuff pressure are sensed by a trans-

ducer inside the instrument. Pairs of oscillations and corre-

sponding cuff pressures are stored electronically to permit

measurement of the systolic and diastolic pressures. Use of

these automated devices is limited in those with irregular

rhythms or when motion cannot be minimized. In addition,

measurements tend to be unreliable in low-flow states.

4. Plethysmography—Arterial pulsations produce minute

changes in the volume of an extremity. Such alterations in

finger volume can be detected photometrically with a

plethysmograph. These devices tend to be less accurate than

alternative pressure monitoring techniques, particularly dur-

ing low-flow and stress conditions.

5. Doppler—The Doppler principle states that any moving

object in the path of a sound beam will alter the frequency of the

transmitted signal. The sound beam used to “insonate” tissue is

created by applying an electrical potential to a crystal that causes

it to oscillate in the radiofrequency spectrum. This sound is cou-

pled to the tissue of interest through an acoustic gel.

When the beam strikes moving blood cells, the frequency

of the reflected beam is altered in a manner proportionate to

the velocity of the reflecting surface. Continuous- and pulsed-

wave Doppler equipment is currently available. Continuous-

wave transducers have two crystals mounted together in a

single probe. One is continuously transmitting, and the other

is continuously receiving. Only the velocity of flow and its

direction can be determined by a continuous-wave device.

Because a Doppler shift occurs only when blood moves rela-

tive to the transducer, an angle correction must be applied:

where Δf is the frequency shift, fe is the frequency of the

insonating beam, V is blood velocity,

is the incident angle

of insonation, and C is the velocity of sound in tissue.

The depth of tissue penetration by the sound beam is

inversely proportional to the frequency of insonation.

Because arteries of interest are typically superficial, a 10-

MHz probe can be used. As can be seen from the equation,

the largest frequency shift is obtained when the probe is held

parallel to the artery. Perpendicular positioning decreases the

frequency shift (cos

→ 0). Doppler blood pressure meas-

urements are obtained by placing an ultrasonic probe on an

artery distal to a compressing cuff.

Doppler sounds become apparent when cuff pressure falls

below arterial pressure. Arterial pressures obtained using a

Doppler probe usually are higher than those obtained by pal-

pation and lower than those obtained by direct measurement,

although the overall correlation is excellent. An automated

device (Arteriosonde) is available for Doppler measurements.

It uses a 2-MHz insonation frequency directed at the brachial

artery. Overall accuracy is very good—especially at low pres-

sures, when ultrasonic and palpatory techniques are more

accurate than auscultation. Disadvantages include motion

sensitivity, requirement for accurate placement, and the need

to use a sonic transmission gel.

B. Invasive Pressure Monitoring—Insertion of a catheter

into an artery is the most accurate technique for pressure

monitoring. Such catheters are connected by tubing to pres-

sure transducers that convert pressure into electrical signals.

Because arterial pressure waves are themselves too weak to

generate electrical impulses, most transducers actually meas-

ure the displacement of an internal diaphragm. This

diaphragm is connected to a resistance bridge such that

motion of the diaphragm modulates an applied current. The

transducer’s sensitivity is the change in applied current for a

given pressure change.

Because transducers are ultimately mechanical, they

absorb energy from the systems they monitor. If absorbed

energy in the transducer’s diaphragm is suddenly released, it

will begin to vibrate at its natural (resonant) frequency. The

tendency for this oscillation to stop depends on the damping

of the system. Oscillating frequency increases as damping

decreases. The resonant frequency is a function of the natu-

ral frequency and the damping coefficient. Classically, a sys-

tem’s damping coefficient is determined by applying and

releasing a square pressure wave (Figure 8–3).

Damping increases when compliance increases. Soft

(compliant) connecting tubing absorbs transmitted pressure

waves and damps the system. Other factors that increase

damping include air in the transducer dome or tubing, exces-

sively long or coiled tubing, connectors containing

diaphragms, and the use of stopcocks. Because air is more

compressible than water, even small bubbles increase the sys-

tem damping. Excessive damping results in underestimation

of systolic pressure and overestimation of diastolic pressure.

There is little effect on mean pressure. Underdamped systems

produce the opposite effects. Additionally, systems with

2feV

(cos )



INTENSIVE CARE MONITORING

191



Figure 8–3. The amplitude ratio obtained by measuring the amplitude of oscillations after pressure release. Either

the listed formula or the tables then can be used to calculate the damping coefficient. (Reproduced, with permission,

McGraw-Hill Companies, Inc.)



CHAPTER 8

192

insufficient compliance tend to “ring” when rapid pressure

changes cause oscillations within the system. Conversely,

overdamping decreases the frequency response to the point

that rapid changes in pressure may not occur. The effect of

damping on the natural frequency of the system is illustrated

in Figure 8–4. The optimal damping coefficient is near 0.7

because there is essentially no effect on amplitude until the

measured frequency approaches the natural frequency of the

measuring system.

Clinical Applications

The arteries commonly used for invasive blood pressure

monitoring, in order of usual preference, are the radial,

ulnar, dorsalis pedis, posterior tibial, femoral, and axillary

arteries. The radial artery is preferred because of its ease of

cannulation and relatively low incidence of serious compli-

cations. The ulnar artery is the dominant artery to the hand

in 90% of patients. It connects with the radial artery through

the palmar arches in 95% of patients. Because vascular

insufficiency may result from occlusion of the dominant

artery, all patients should undergo an Allen test prior to

catheter insertion, and the results should be entered into the

medical record. However, one prospective study has demon-

strated that vascular complications were not reliably related

to results of the Allen test. Overall, there is a 10% incidence

of arterial occlusion in adults cannulated with 20-gauge

Teflon catheters for a period of 1–3 days. The use of 22-gauge

catheters seems to reduce this incidence.

For unknown reasons, women have a lower incidence of

arterial thrombosis than men. When thromboses do occur in

women, occlusions are usually temporary. Distal occlusion of

the radial artery may cause overestimation of systolic pres-

sure because of increased wave reflection, whereas proximal

occlusion usually causes reduction in pressure owing to

overdamping. Another complication of arterial catheters is

infection, most commonly limited to the skin but some-

times involving the artery as well; distal septic emboli rarely

occur. The incidence and severity of such infections can be

minimized by strict adherence to policies of daily catheter



Figure 8–4. The amplitude ratio depends on the natural frequency of the system and the damping ratio (h). A sys-

tem that operates below its natural frequency and with a damping ratio near 0.7 is desirable. (Reproduced, with

permission, from Fry DL: Physiologic recording by modern instruments with particular reference to pressure recording.

Physiol Rev 1960;40.)



INTENSIVE CARE MONITORING

193

inspection, sterile dressing changes, and limiting catheteriza-

tion to 5 days or less at any single site. Pseudoaneurysm for-

mation may be a late complication of arterial catheters. The

incidence of pseudoaneurysms may be minimized by using

smaller catheters, minimizing the duration of catheteriza-

tion, and preventing catheter infections.

Physiologic pressures are measured with reference to the

tricuspid valve, where intravascular pressure is defined as

zero. This phlebostatic axis is independent of changes in

body habitus. Postural changes affect the reference pressure

by less than 1 mm Hg. The phlebostatic point is identified as

(1) 61% of the way from the back to the front, (2) exactly in

the midline, and (3) one-quarter of the distance above the

inferior tip of the xiphoid process. A convenient method of

system calibration is to move an open stopcock attached by

fluid-filled connecting tubing to the transducer up against

the patient’s midaxillary line. The digital display on the mon-

itor indicates whether the midaxillary line is above (positive

pressure) or below (negative pressure) the transducer. The

bed is then moved up or down until the pressure reads zero

(Figure 8–5).

Calibration of the monitor for nonzero pressure can be

done internally or externally. External calibration can be

done with a mercury manometer for systemic arterial pres-

sures. A convenient method adequate for the lower pressure

range required for a pulmonary artery catheter (up to 60 cm

O) takes advantage of the fluid-filled connecting tubing.

After establishing the zero reference, move an open stopcock

connected to the transducer above the transducer a meas-

ured amount. The height above the transducer in centimeters

(cm H

O pressure) should be read by the monitor system in

millimeters of mercury as height in centimeters divided

by 1.36. Therefore, if the system is calibrated accurately, the

pressure reading should be about 14.8 mm Hg when the

stopcock is raised 20 cm above the transducer.

An additional use of arterial catheterization is to provide

access for arterial blood sampling. This is often indicated in

patients who require frequent sampling of blood for arterial

blood gases or other blood tests.

Araghi A, Bander JJ, Guzman JA: Arterial blood pressure monitor-

ing in overweight critically ill patients: Invasive or noninvasive?

Crit Care 2006;10:R64. [PMID: 16630359]

Bur A et al: Factors influencing the accuracy of oscillometric blood

pressure measurement in critically ill patients. Crit Care Med

2003;31:793–9. [PMID: 12626986]

Gibbs NC, Gardner RM: Dynamics of invasive pressure monitor-

ing systems: Clinical and laboratory evaluation. Heart Lung

1988;17:43–51. [PMID: 3338943]

Gunn SR, Pinsky MR: Implications of arterial pressure variation in

patients in the intensive care unit. Curr Opin Crit Care

2001;7:212–7. [PMID: 11436530]

Mignini MA, Piacentini E, Dubin A: Peripheral arterial blood pres-

sure monitoring adequately tracks central arterial blood pres-

sure in critically ill patients: An observational study. Crit Care

2006;10:R43. [PMID: 16542489]

Pittman JA, Ping JS, Mark JB: Arterial and central venous pressure

monitoring. Int Anesthesiol Clin 2004;42:13–30. [PMID:

14716195]



Central Venous Catheters

Central venous (CV) catheters are inserted via the subcla-

vian, internal jugular, or a peripheral vein in the arm.

Femoral venous catheters are not long enough to reach

“central” veins but provide similar access for intravenous



Figure 8–5. The height of the transducer must be adjusted to the phlebostatic axis to ensure accuracy of the

pressure measurements.



CHAPTER 8

194

infusions. For monitoring purposes, CV catheters provide

estimates of central venous pressure (CVP) and measure-

ment of central venous oxygen saturation (Scv

). Central

venous pressure (CVP) reflects the balance between systemic

venous return and cardiac output. In the normal heart, the

right ventricle is more compliant than the left. This differ-

ence in compliance accounts for the slope of their correspon-

ding Frank-Starling curves. The use of CVP to assess

left-sided preload causes difficulty because CVP primarily

reflects changes in right ventricular end-diastolic pressure

and only secondarily reflects changes in pulmonary venous

and left-sided pressures. The relationship between CVP and

venous return is shown in Figure 8–6A. Decreasing right

atrial pressure below zero does not significantly increase

CVP because of collapse of the vasculature leading to the

chest. The figure also demonstrates that changes in mean sys-

temic pressure cause a parallel change in venous return.

Alterations in vascular resistance (decreased by anemia, arte-

riovenous fistulas, pregnancy, or thyrotoxicosis) change the

slope of the respective curves (Figure 8–6B).

A water manometer may be used to measure CVP. The

normal range of CVP is between –4 and +10 mm Hg (–5.4

and +13.6 cm H

O).

An electronic transducer also displays the pressure wave-

form. The bandwidth of a catheter-transducer system used to

monitor CVP can be significantly narrower than that used

for arterial pressure.

A typical CVP waveform has three positive deflections

(a, c, and v) and two descents (x and y) (Figure 8–7). The

increase in venous pressure caused by atrial contraction pro-

duces the a wave. The c wave is created when the tricuspid valve

is displaced into the right atrium during initial ventricular

contraction. The x descent corresponds to the period of ven-

tricular ejection, when blood empties from the heart; it is

inscribed when the ventricle draws down on the floor of the

atrium and decreases the CVP. The v wave is produced by the

increase in atrial pressure that takes place as venous return

continues while the tricuspid valve is closed. The y descent

occurs when the tricuspid valve opens at the conclusion of

ventricular contraction, and blood enters the right ventricle.

The a wave is absent during atrial fibrillation and is magni-

fied by tricuspid stenosis (cannon wave). The x descent also

may be absent with atrial fibrillation. The x and y descents

are both exaggerated by constrictive pericarditis. Cardiac

tamponade magnifies the x descent while abolishing the y

descent. When tricuspid regurgitation occurs, the c wave and

the x descent are replaced by a large single regurgitant wave.

Pulmonary hypertension decreases right ventricular compli-

ance and accentuates the v wave.

Clinical Applications

A. CVP Monitoring—CVP monitoring is best used for

patients without preexisting cardiac disease as one indicator

of the adequacy of venous return and cardiac filling. An

intravenous fluid challenge is employed to aid in determin-

ing whether decreased blood pressure is due to hypovolemia

or to cardiogenic failure. Measurements of CVP are affected

by ventilation because transthoracic pressure is transmitted

through the pericardium and the thin-walled venae cavae.

During spontaneous ventilation, inspiration lowers CVP, and

exhalation increases it. The situation is reversed in patients

being mechanically ventilated, in whom inspiration increases

intrathoracic pressure and elevates CVP.

Venous return (L/min)



Figure 8–6.

Effect of mean systemic pressure on venous return.

Effect of systemic vascular resistance on

venous return. (Reproduced, with permission, from Otto CW: Central venous pressure monitoring. In: Blitt CD (ed),



INTENSIVE CARE MONITORING

195

The degree of this elevation depends on the compliance

of the lungs and intravascular volume and will vary among

patients. For this reason, CVP measurements are best made

and compared at end expiration. When positive end-

expiratory pressure (PEEP) is applied, the positive pressure is

transmitted through to the right atrium, causing a decrease

in venous return and a rise in CVP. Again, the magnitude of

this effect of PEEP on CVP varies with pulmonary compli-

ance and blood volume. Some argue that the patient should

be temporarily removed from PEEP while the measurement

is taken. This is both impractical and potentially dangerous.

In critical situations, an esophageal probe can be inserted to

estimate transthoracic pressure. Subtracting the transtho-

racic pressure from the CVP provides transmural pressure,

which is a better estimate of right atrial pressure in the pres-

ence of elevated transthoracic pressure.

B. Central Venous O

Saturation—Mixed venous oxygen

saturation (S

–

) reflects the relative delivery of O

to the tis-

sues compared with consumption. If lower than normal,

concern should be raised about tissue hypoxia. True S

–

must be measured in the pulmonary artery. Central venous

saturation (Scv

) does not require a pulmonary artery

catheter, but theoretically, the value will differ from S

–

because Scv

obtained from a subclavian or internal jugular

vein does not reflect venous blood returning via the inferior

vena cava or coronary sinus. Generally, Scv

is about 5%

higher than S

–

In practice,however,Scv

appears to have similar predic-

tive value for end-organ hypoxia as S

–

. Recent studies

emphasizing early goal-directed therapy in sepsis have

emphasized a target Scv

of greater than 70% by giving

blood transfusion and cardiac inotropic drugs. Scv

can be

obtained from a small sample of blood drawn back through

the catheter or by using an oximeter-tipped CVP catheter.

C. Complications—Inadvertent arterial insertion occurs

about 2% of the time; such insertion is particularly dangerous

if large, rigid “introducer” catheters are inserted. Perforation

of the superior vena cava is associated with a 67% mortality

rate, whereas the rate associated with laceration of the right

ventricle approaches 100%. Such perforations may occur

either from guidewires or from catheter erosion—again, par-

ticularly with introducer catheters. Other structures that may

be injured on insertion include the brachial plexus, the stel-

late ganglion, and the phrenic nerve. Air emboli are uncom-

mon at insertion but more often during use or at the time of

removal when the patient is not positioned properly. Late

complications are due to catheter migration, embolization,

and infection. The incidence of cannula-related thrombosis

of the axillary and subclavian veins varies between 16.5% and

46%. Central venous catheter infections occur in approxi-

mately 5% of insertions. The organisms most commonly

involved are Staphylococcus epidermidis, 30%; Staphylococcus

aureus, 8%; streptococci, 3%; gram-negative rods, 18%; diph-

theroids, 2%; Candida species, 24%; and other pathogens, 15%.

Both colonization of central venous catheters and systemic

sepsis are reduced by routine catheter care and periodic

removal and reinsertion.

Kalfon P et al: Comparison of silver-impregnated with standard

multi-lumen central venous catheters in critically ill patients.

Crit Care Med 2007;35:1032–9. [PMID: 17334256]

Kusminsky RE: Complications of central venous catheterization.

J Am Coll Surg 2007;204:681–96. [PMID: 17382229]

Marx G, Reinhart K: Venous oximetry. Curr Opin Crit Care

2006;12:263–8. [PMID: 16672787]

Michard F, Teboul JL: Predicting fluid responsiveness in ICU patients:

A critical analysis of the evidence. Chest 2002;121:2000–8. [PMID:

12065368]

Onders RP, Shenk RR, Stellato TA: Long-term central venous

catheters: Size and location do matter. Am J Surg 2006;

191:396–9. [PMID: 16490554]

Pinsky MR, Teboul JL: Assessment of indices of preload and vol-

ume responsiveness. Curr Opin Crit Care 2005;11:235–9.

[PMID: 15928472]

Rivers E, et al. Early goal-directed therapy in the treatment of

severe sepsis and septic shock. N Engl J Med. 2001;345:1368–77.

[PMID: 11794169]

Rivers E: Mixed vs central venous oxygen saturation may be not

numerically equal, but both are still clinically useful. Chest

2006;129:507–8. [PMID: 16537845]

Inspiration Inspiration

Water

manometer

pressure

Mean end expiratory pressure

ExhalationExhalation

Right atrial

pressure (mm Hg)



Figure 8–7. Effect of mechanical ventilation on central venous pressure. The

a, c

, and

waves along with the

and

descents are shown. (Reproduced, with permission, from Otto CW: Central venous pressure monitoring. In: Blitt CD



CHAPTER 8

196

Rivers EP, Ander DS, Powell D: Central venous oxygen saturation

monitoring in the critically ill patient. Curr Opin Crit Care

2001;7:204–211. [PMID: 11436529]

Taylor RW, Palagiri AV: Central venous catheterization. Crit Care

Med 2007;35:1390–6. [PMID: 17414086]



Pulmonary Artery Catheterization

Catheterization of the pulmonary artery is a useful addition

to CVP monitoring. It provides information related to left

heart filling pressures and allows sampling of pulmonary

artery blood for determination of mixed venous oxygen sat-

uration. Thermodilution cardiac output measurements are

made using a thermistor-tipped catheter.

As the balloon flotation catheter is advanced through the

heart, characteristic pressure waveforms are obtained that

indicate the position of the catheter’s distal port (Figure 8–8).

Simultaneous electrocardiographic monitoring ensures that

ventricular tachyarrhythmias will be detected as the

catheter traverses the right ventricle. After a pulmonary

capillary wedge tracing is obtained, the catheter should be

deflated and withdrawn until only 1 mL of inflation is

required to advance from a pulmonary artery to a capillary

wedge tracing. Insertion of excessive catheter length con-

tributes to intracardiac knotting. If subsequent pressure

tracings are not obtained within 15 cm of additional inser-

tion, looping should be suspected. When the catheter is

placed through either the subclavian or the jugular vein, the

typical distances required are as follows: right atrium, 10–15

cm; right ventricle, 20–30 cm; pulmonary artery, 45–50 cm;

and pulmonary capillary wedge, 50–55 cm. As the catheter

passes through the right ventricle, a wedgelike pressure trac-

ing may be obtained. This “pseudowedge” is due to engage-

ment of the catheter tip beneath the pulmonary valve or

within trabeculations. Withdrawal of 10 cm of the catheter

will solve the problem. Overinflation of the balloon, causing

it to herniate over the tip of the catheter, results in a pressure

tracing that continues to rise to high levels. The balloon

should be deflated and a short length of catheter withdrawn

before further advancement is attempted.

The final position of the catheter tip within the pul-

monary artery is critical. This may be described with refer-

ence to three lung zones that depend on the relationship of

airway and vascular pressures (Figure 8–9). In zones I and II,

mean airway pressure is intermittently greater than pul-

monary venous pressure, which results in collapse of the vas-

culature between the catheter tip and the left atrium. In this

position, observed pressures will be more indicative of air-

way pressure than of left atrial pressure. Only in zone III is

there an uninterrupted column of blood between the

catheter and the left atrium. In the supine position, zone III

assumes a more dependent position, caudal to the atrium.

Decreased airway pressures change the ventilation-perfusion

relationship, producing a relative increase in zone III.

Hypovolemia decreases vascular pressures and decreases

zone III.

Correct catheter position should be ensured by chest x-

ray. Although most catheters migrate caudally and to the

right, an occasional catheter will become wedged anterior to

the vena cava. In this position, true pulmonary capillary

pressures may be less than alveolar pressures, resulting in

spuriously elevated measurements. A lateral chest x-ray will

mm Hg



Figure 8–8. Normal pressures and waveforms obtained as a pulmonary artery flotation catheter is advanced from

the right atrium to a pulmonary artery wedge position. (Reproduced, with permission, from Morgan GE, Mikhail MS: