Pulmonary Artery Catheterization Technique

Zero reference

Any independent vertical movement of the transducer or the patient will affect the hydrostatic column of this fluid-filled system and thus alter the pressure measurements. At some time before or after PAC insertion, the system must therefore be zeroed to ambient air pressure.

The reference point for this is the midpoint of the left atrium (LA), estimated as the fourth intercostal space in the midaxillary line with the patient in the supine position. With the transducer at this height, the membrane is exposed to atmospheric pressure, and the monitor is then adjusted to zero.

Calibration

Once zeroed, the monitoring system must be calibrated for accuracy. Currently, most monitors perform an automated electronic calibration.

Two methods are used to manually calibrate and check the system, as follows:

• If the catheter has not been inserted, the distal tip of the PAC is raised to a specified height above the LA. For example, raising the tip 20 cm above the LA should produce a reading of approximately 15 mm Hg if the system is working properly (1 mm Hg equals 1.36 cm H₂ O).

• Alternatively, pressure can be applied externally to the transducer and adjusted to a known level using a mercury or aneroid manometer. The monitor then is adjusted to read this pressure, and the system is calibrated.

Dynamic tuning

Central pressures are dynamic waveforms (ie, they vary from systole to diastole) and thus have a periodic frequency. To monitor these pressures accurately, the system requires an appropriate frequency response. A poorly responsive system produces inaccurate pressure readings, and differentiating waveforms (eg, PA from pulmonary capillary wedge pressure [PCWP]) can become difficult. When signal energy is lost, the pressure waveform is dampened. Common causes of this are air bubbles (which are compressible), long or compliant tubing, vessel wall impingement, intracatheter debris, transducer malfunction, and loose connections in the tubing. A qualitative test of the frequency response is performed by flicking the catheter and observing a brisk high-frequency response in the waveform.

After insertion, the system can be checked by using the rapid flush test. When flushed, an appropriately responsive system shows an initial horizontal straight line with a high-pressure reading. Once the flushing is terminated, the pressure drops immediately, which is represented by a vertical line that plunges below the baseline. A brief and well-defined oscillation occurs, followed by return of the PA waveform. A dampened system will not overshoot or oscillate, and causes a delay in returning to the PA waveform.

The PAC is inserted percutaneously into a major vein (jugular, subclavian, femoral) via an introducer sheath. The actual venous access techniques are not described here, but the following points are important. Preference considerations for cannulation of the great veins are as follows:

• Right internal jugular vein (RIJ) - Shortest and straightest path to the heart

• Left subclavian - Does not require the PAC to pass and course at an acute angle to enter the SVC (compared to the right subclavian or left internal jugular [LIJ])

• Femoral veins - These access points are distant sites, from which passing a PAC into the heart can be difficult, especially if the right-sided cardiac chambers are enlarged. Often, fluoroscopic assistance is necessary. Nevertheless, these sites are compressible and may be preferable if the risk of hemorrhage is high.

As with any catheterization procedure, sterile technique is essential. The total length of a PAC is approximately 150 cm; extra sterile towels around the head, shoulders, and chest ensure that aseptic technique is not compromised.

While the Trendelenburg position is used for venous access (internal jugular [IJ] and subclavian routes), passage of the PAC is easier when the patient subsequently is placed flat or slightly upright.

Before insertion, check the PAC for cracks and kinks. Then, check balloon function (see image below), connect all lumens to stopcocks, and flush them to eliminate air bubbles. Flick the PAC tip to check frequency response. Finally, the PAC is threaded through a sterile sleeve (be sure to check orientation) to ensure sterility of the PAC after insertion and allow some adjustment of position.

The packaging of the PAC causes it to have a preformed curve. This can be used to facilitate passage into the PA. The direction in which the curl is inserted into the introducer depends on which vein is cannulated. For instance, from the head of the bed using the RIJ approach, the curl should be in the direction of the patient's left shoulder (concave-cephalad). Once the PAC is in the RV, a clockwise quarter turn moves the tip anteriorly to allow easier passage into the PA.

After inserting the PAC as far as the 20-cm mark (30-cm mark if the femoral route used), the balloon is inflated with air. Inflation should be slow and controlled (1 mL/s) and should not surpass the recommended volume (usually 1.5 mL). Always inflate the balloon before advancing the PAC, and always deflate the balloon before withdrawing the PAC.

Always use continuous pressure monitoring from the distal lumen. Watch the monitor for changes in the waveform and abnormal cardiac rhythms. From the RIJ approach, the RA is entered at approximately 25 cm, the RV at approximately 30 cm, and the PA at approximately 40 cm; the PCWP can be identified at approximately 45 cm.

If an RV waveform still present approximately 20 cm after the initial RV pattern appears, the catheter may be coiling in the RV. If withdrawal is necessary, always proceed slowly to decrease the risk of knotting the catheter upon itself. If the catheter is knotted, fluoroscopy may be necessary to visualize the catheter and remove the knot. As a last resort, slowly withdraw the PAC to the point where it catches on the introducer tip. From this point, the PAC and introducer can be removed as one unit. Apply prompt pressure for a minimum of 5 minutes. If bleeding persists, suturing the site may be necessary.

Once the PCWP is obtained and the catheter sleeve secured, make sure the PCWP pattern is reproducible before removing the sterile field. Also, determine the volume of air in the balloon required to obtain a PCWP waveform. Volumes less than half the balloon maximum may indicate that the tip is too far distal. Some clinicians advocate that, after establishing that the PA diastolic pressure is equal to the PCWP pressure, further balloon inflations are unnecessary and the PA diastolic pressure should be used as the parameter to assess left ventricular (LV) filling; this relationship may not hold if the clinical situation changes.

Once the procedure is complete, obtain a chest radiograph to check the position of the PAC and to assess for central venous access complications (eg, pneumothorax).

An interesting real-time online video is available to enhance the visualization of the course of the catheter as it passes through the heart. ^{[34, 35, 36]}

Table 2 below shows the normal range of pressures for the RA, RV, PA, and PCWP.

Other important information provided by a PAC catheter includes the CO, mixed venous oxygen saturation (SaO₂), and oxygen saturations in the right heart chambers to assess for the presence of an intracardiac shunt.

Using these measurements, other variables can be derived, including pulmonary or systemic vascular resistance and the difference between arterial and venous oxygen content (see the image and Table 1 below). Obtaining CO and PCWP measurements is the primary reason for inserting most PACs; therefore, understanding how they are obtained and what factors alter their values is of prime importance.

 

Table 1. Calculation of Various Hemodynamic Parameters (Open Table in a new window)

Table 2. Circulatory Pressures (Open Table in a new window)

Wave form analysis in healthy states

Cardiac pressures

Right- and left-sided heart pressure waveforms share many physiologic similarities, but, in the healthy individual, the waves are of different magnitudes.

Right-sided pressures

The central venous pressure (CVP) and right atrial pressure (RAP) are nearly equal to the diastolic RV pressure in the absence of heart or lung disease (see image below).

The mean CVP and RAP normally range from 0-5 mm Hg, and vary as intrathoracic pressure changes with respiration (see image below).

RA contraction creates pressure changes, which are influenced strongly by the patient's volume status. Atrial contraction produces an increase in pressure called the A wave. The C wave is a small convexity noted on the initial descent of the A wave and is thought to be secondary to closure of the tricuspid valve. The initial descent after the A wave is called the X descent. This decline in RAP is secondary to RA relaxation and downward movement of the tricuspid valve. Following this is the V wave, which is somewhat smaller than the A wave, and reflects RA filling during ventricular systole. The Y descent occurs after the V wave and represents rapid filling of the RV after opening of the tricuspid valve (see image below).

CVP is most commonly elevated in the setting of biventricular heart failure. Other causes of RAP elevation are tricuspid regurgitation or stenosis, pulmonary hypertension, volume overload, constrictive pericarditis, and cardiac tamponade. Large, so-called cannon A waves occur when the RA contracts against a closed tricuspid valve. Cannon A waves are detected in certain cardiac rhythm disturbances, including junctional rhythms and ventricular tachycardia, and in some patients with ventricular pacemakers. Large V waves may occur in the presence of tricuspid regurgitation, with their magnitude affected by the size and compliance of the RA.

Pulmonary arterial pressure

In the pulmonary artery pressure (Ppa) tracing, an initial positive upstroke secondary to RV systole occurs, and a dicrotic notch is formed on the downstroke when the pulmonary valve closes. A normal PA systolic pressure ranges from 20 to 30 mm Hg and is equal to the RV systolic pressure. Ppa is elevated in some high-flow states (eg, hypervolemia), left ventricular failure, and high-resistance states (eg, pulmonary hypertension, mitral valve disease) (see image below).

Pulmonary artery occlusion pressure (wedge pressure)

Understanding the theory and required assumptions behind PCWP measurement and conditions that alter it are essential for proper use of this often misunderstood measurement. When the PAC tip is positioned properly and the balloon is inflated, the PAP tracing disappears. This occurs because inflation of the balloon causes distal migration (approximately 2 cm) of the tip into a smaller branch of the PA, where it occludes blood flow. The resulting nonpulsatile pressure tracing is called the PCWP (or pulmonary artery occlusion pressure [Ppao]) (see image below).

Under the proper circumstances, this pressure reflects the mean left atrial pressure (LAP) (see image below).

The assumption is that a static column is created between the PAC tip and the LA. This assumption is correct only if the tip is in the proper lung zone and no vascular obstruction, such as pulmonary vein stenosis, occurs downstream. When the PAC catheter balloon is inflated, the balloon stops antegrade blood flow and allows an uninterrupted column of blood to exist between the catheter tip and the LA (see images below).

The PCWP waveform reflects events in the LA. The A, C, and V waves have origins similar to those that appear in the RAP waveform (see image below).

The waveforms can be discerned by using simultaneous ECG monitoring (see image below).

The 3 lung zones of West

The lung can be divided into 3 vertical zones with varying pressure changes (see image below). ^[37]

In zone 1 (apex), alveolar pressure (Palv) exceeds both mean Ppa and pulmonary venous pressures (Ppv). Flow depends on Palv. In zone 2 (central), Ppa is greater than Palv, which is greater than Ppv, and flow depends on a balance between Ppa and Palv. Because capillary collapse is present, neither zone 1 nor zone 2 allows a direct connection with the LA. In zone 3 (lung bases), Palv is less than Ppa and Ppv. Flow is not interrupted, and a direct column of blood extends to the LA.

Fortunately, the actual practice of placing the tip in zone 3 to ensure more accurate measurements of LAP is not complicated. In the supine patient, most of the lung is considered zone 3. Blood flow to this area is increased, making balloon flotation easier. In critically ill patients who require positive end expiratory pressure (PEEP) levels greater than 10 cm H₂ O, the zone 3 area can be reduced.

To assess proper location, a supine chest radiograph showing the tip below the level of the LA is sufficient, although occasionally a lateral chest radiograph is required. If the tip position remains questionable, blood can be aspirated from the distal port during balloon inflation.

Preload (left ventricular end-diastolic pressure)

PCWP is a reflection of LAP, which, in the absence of mitral valve disease, is an indication of LV diastolic pressure. Often, the inference is made that PCWP reflects left ventricular end-diastolic volume (LVEDV) or end-diastolic pressure (LVEDP). Numerous conditions in critically ill patients preclude this assumption.

PCWP is the measurement by which changes in lung water (pulmonary capillary hydrostatic pressure [PCHP]) can be assessed. This concept holds true only if the resistance of the pulmonary venous system is assumed to be zero. In fact, the small pulmonary veins and capillaries account for approximately 40% of the total pulmonary vascular resistance. This value may be even higher in critically ill patients in whom pulmonary venoconstriction is common secondary to conditions such as hypoxemia and acute respiratory distress syndrome (ARDS). PCHP is always greater than PCWP. PCWP can be used to estimate the contribution of PCHP to lung edema if evidence of chronically elevated Ppv, permeability, pleural pressure, and osmotic pressure are considered.

Effect of respiration

The final critical concept in PCWP interpretation is the effect of the respiratory cycle on PCWP measurements. The timing of PCWP measurement is critical because intrathoracic pressures can vary widely with inspiration and expiration and are transmitted to the pulmonary vasculature. During spontaneous inspiration, the intrathoracic pressures decrease (more negative); during expiration, intrathoracic pressures increase (more positive). Positive pressure ventilation (eg, in an intubated patient) reverses this situation. To minimize the effect of the respiratory cycle on intrathoracic pressures, measurements are obtained at end-expiration, when intrathoracic pressure is closest to zero.

In patients with severe respiratory distress, end-expiration can be difficult to determine. In these situations, sedation, or even paralysis, may be necessary to remove the transmission of respiratory efforts to the pressure tracings.

Positive end-expiratory pressure

PEEP (intrinsic or extrinsic) also transmits pressure to the vascular space. Lung compliance is the main determinant of the amount of pressure transmission. For example, in disease states (eg, ARDS) associated with low compliance (ie, stiff lungs), pressure transmission is minimal. Debate exists over how to correct PCWP in the presence of PEEP. Although previously advocated, temporary discontinuation of PEEP may have adverse effects, such as cardiovascular collapse or hypoxemia, that are difficult to reverse.

For PEEP greater than 10 cm H₂ O, the following general rule can be applied: Corrected PCWP equals measured PCWP minus one half the quotient of PEEP divided by 1.36. If available, an intraesophageal balloon can be used. Esophageal pressure equals pleural pressure, so corrected PCWP equals measured PCWP minus esophageal pressure.

Wave form analysis in pathologic states

Shock has been defined as inadequate perfusion to meet the metabolic demands of body tissues. The most common forms of shock are hypovolemic, cardiogenic, septic, and obstructive. PACs are used frequently in the management of various forms of shock, as described in this section (see image below).

Hypovolemic shock

Hypovolemic shock is due to a reduction in circulating blood volume resulting from either hemorrhage or fluid depletion. Preload is markedly decreased, leading to inadequate ventricular filling. The patient with hypovolemic shock manifests hypotension and tachycardia. Systemic, venous, and intracardiac pressures are abnormally low. The overall PAC pressure tracing has a damped appearance.

Cardiogenic shock

Cardiogenic shock is the result of severe depression in cardiac performance. Cardiogenic shock is characterized by systolic blood pressure less than 80 mm Hg, cardiac index less than 1.8 L/min/m², and PCWP greater than 18 mm Hg. This form of shock can occur from a direct insult to the myocardium (eg, large AMI, severe cardiomyopathy) or from a mechanical problem that overwhelms the functional capacity of the myocardium (eg, acute severe mitral regurgitation, acute ventricular septal defect).

Common causes of acute mitral regurgitation in critical care units are ruptured papillary muscles secondary to AMI, myocardial ischemia leading to papillary muscle dysfunction, bacterial endocarditis, ruptured chordae, and trauma. Other causes are rheumatic fever and myxomatous degeneration of the mitral valve. With acute mitral regurgitation, large volumes of blood regurgitate into a poorly compliant LA, raising Ppv and causing pulmonary edema.

Large V waves usually are observed in the PCWP pressure tracing (see images below).

The PA waveform appears falsely elevated because of the large V wave reflected back from the LA through the compliant pulmonary vasculature. The Y descent is quite rapid as the overdistended LA quickly empties. Care must be exercised to distinguish a large V wave from a systolic PA waveform. Failure to recognize a large V wave may cause the PAC to be advanced further in an attempt to record a PCWP pressure, increasing the risk of perforation.

In chronic mitral regurgitation, an equivalent volume of blood may regurgitate, but this volume is better tolerated by a markedly dilated LA. Compared with acute mitral regurgitation, LA pressure may be less and large V waves may be absent.

Septic shock

Septic shock is the most common cause of death in intensive care units in the United States. Septic shock is an example of distributive shock, a form of shock characterized by profound peripheral vasodilation. Although the CO may be normal or even elevated in this type of shock, organ and tissue perfusion are inadequate. Other types of distributive shock include anaphylaxis, neurogenic shock, and adrenal insufficiency. Swan-Ganz catheter measurements frequently demonstrate low filling pressures.

Extracardiac obstructive shock

Pericardial tamponade is an example of this form of shock. Cardiac tamponade results from abnormal rapid fluid accumulation in the pericardial sac. The increased pericardial pressure impairs ventricular diastolic filling, decreasing preload, stroke volume, and CO. This may occur secondary to viral infections, malignancy, trauma, or myocardial rupture. As little as 50 mL of fluid accumulation can begin to impair cardiac filling during systole, leading to a severe reduction in CO. Ventricular filling is impaired throughout all of diastole, thereby causing equalization of all diastolic pressures.

The RAP approximates the RV diastolic pressure, which approximates the PA diastolic pressure, and also approximates PCWP (see image below).

The RA waveform shows a minimal X and small and/or absent Y descent, and the mean RAP is elevated. Ppa loses its usual respiratory variation. In pericardial tamponade, the systemic arterial pressure shows evidence of pulsus paradoxus (see image below). Other causes of extracardiac shock include massive PE and tension pneumothorax.

Hemodynamics of other cardiac abnormalities

Constrictive pericarditis

Thickening of the pericardial sac creates an indolent process that may lead to constrictive pericarditis. This can occur in patients with rheumatic diseases, tuberculosis, metastatic cancer, or prior chest radiation or open-heart surgery. Idiopathic cases also occur. Early diastolic filling is normal until limited by the rigid pericardial shell. Once this occurs, ventricular filling is stopped abruptly, creating a plateau in the RV pressure, which is typical of constrictive pericarditis. This is called the "dip and plateau" pattern or square root sign; the RAP waveform has a characteristic configuration suggestive of an M or W. A and V waves are accentuated with rapid X and Y descents, in contrast to pericardial tamponade, above. PCWP may be as high as 20-25 mm Hg, and usually appears similar to the RA waveform. Pulsus paradoxus is present much less commonly with constrictive pericarditis than with pericardial tamponade (see images below).

Mitral stenosis

In severe mitral stenosis, LAP, and thus PAWP, is elevated. Pulmonary hypertension also develops as the severity of the valve lesion progresses. This leads to increase in RV systolic pressure and in the RA A wave. RV diastolic pressure may increase if RV failure or important tricuspid regurgitation develops. Atrial fibrillation is a common complication in mitral stenosis and results in loss of A waves in both the RA and PCWP pressure tracings.

Aortic stenosis

Aortic stenosis can be supravalvular, valvular, or subvalvular in origin. The RA, RV, and PA waveforms usually are normal unless congestive heart failure is present. PCWP may show large A waves in severe cases because of poor LV compliance.

Aortic regurgitation

The hemodynamic abnormalities are different in acute and chronic aortic regurgitation. Acute regurgitation is observed most often in bacterial endocarditis, chest trauma, ascending aortic dissection, and degeneration of valve leaflets. The hemodynamics in acute aortic regurgitation include modestly elevated RAP and elevated RV systolic and diastolic pressures. PA systolic and diastolic pressures also are elevated, as is PCWP. A widened and elevated systemic arterial pressure without a dicrotic notch is sometimes observed. Acute and chronic aortic regurgitation often present with contrasting manifestations; a wide pulse pressure usually is not observed in acute regurgitation.

Other hemodynamic measurements made by the pulmonary artery catheter

Cardiac output

CO can be determined via the PAC by several methods. It can be determined by using the Fick principle, which is a variation of the law of conservation and states that consumption of a substance must equal the product of blood flow to the organ and the difference between the arterial and venous concentrations of the substance. In this circumstance, the substance is oxygen, and CO is determined by the following formula:

CO equals oxygen consumption per minute (VO₂) divided by arterial oxygen content (Ca_O2) minus mixed venous oxygen content (Cv_O2)

CO is determined by using systemic arterial and PA blood samples, and by measuring or estimating VO₂. The Fick method is most accurate when the CO is low and the arterial-venous oxygen difference is high. Unfortunately, in critically ill patients, establishing a steady-state and estimating or measuring VO₂ is difficult; thus, the reliability of this technique is poor.

The indicator-dilution technique is more accurate and reproducible. A known amount of dye (indocyanine green) is injected into the PA. Arterial blood is withdrawn from the aorta as the dye circulates, and a concentration-versus-time curve is derived. The first-pass curve is used to determine CO, which is calculated by dividing the initial mass of the injectate by the average concentration. This value then is corrected (60 s/time of the curve) to obtain CO. This procedure requires considerable blood sampling and is time consuming, because recirculation of the dye complicates the calculation.

Many PACs also allow CO to be measured by using a variation of the indicator-dilution method known as the thermodilution method. This method is more efficient because the injectate does not recirculate to a significant degree and no blood sampling is necessary. A saline bolus of known volume (5-10 mL) and temperature (usually ≤25ºC) is injected through the proximal (RA) lumen. The thermistor at the end of the PAC monitors the change in blood temperature, and a temperature-versus-time curve is generated (see image below).

The change in temperature as warm venous blood dilutes the injectate is inversely proportional to the derived CO. The Stewart-Hamilton formula shows this relationship: CO equals the volume of injectate multiplied by blood temperature minus injectate temperature multiplied by computation constants, and divided by change of blood temperature as a function of time (area under the curve).

Understanding this formula allows discernment of artifact errors that can lead to underestimation or overestimation. Loss of injectate or inadvertent administration of a volume lower than required results in a low-amplitude temperature-versus-time curve that produces a falsely elevated CO value. Causes of this are system leak, right-to-left intracardiac shunts, inappropriately rapid injection, and a poorly positioned PAC. Conversely, too much injectate or too slow an injection leads to a falsely low CO reading. Temperature errors can occur when continuous infusions are used. Thrombus or vessel wall impingement can alter the thermistor function.

Physiologic causes for CO measurement discrepancies include tricuspid and pulmonary regurgitation, which may produce recirculation peaks and thus increase the area under the curve, resulting in a falsely low CO estimate. Arrhythmias alter steadiness of PA flow and may cause difficulty in obtaining a consistent CO.

CO alterations occur during the respiratory cycle and are accentuated by respiratory distress and positive pressure ventilation. Proper timing of the injection to the same phase of respiration (preferably end-expiration) provides more consistent measurements. Averaging the values of 3 injections is recommended to minimize sampling errors.

While CO is one of the most important measurements that the PAC provides, the absolute value should be normalized for the size of the patient. To account for this, the cardiac index (CI), which equals CO divided by body surface area [BSA], is calculated. The physician should keep in mind that, as independent variables, CO and CI are of limited use for assessing tissue perfusion because these must be interpreted along with other clinical data.