Pulmonary artery catheterization: Indications, contraindications, and complications in adults

INTRODUCTION — Pulmonary artery catheters (PACs; also called Swan-Ganz catheters) are used for the evaluation of a range of conditions. Although their routine use has fallen out of favor, they are still occasionally placed for the management of critically ill patients, and for the evaluation of unexplained dyspnea or suspected pulmonary hypertension.

The indications, contraindications, and complications of PACs are reviewed here. The insertion of PACs and interpretation of PAC measurements are discussed separately. (See "Pulmonary artery catheters: Insertion technique in adults" and "Pulmonary artery catheterization: Interpretation of hemodynamic values and waveforms in adults" and "Cardiac catheterization techniques: Normal hemodynamics".)

PHYSIOLOGIC MEASUREMENTS — Direct measurements of the following can be obtained from an accurately placed pulmonary artery catheter (PAC):

●Central venous pressure (CVP)

●Right-sided intracardiac pressures (right atrium, right ventricle)

●Pulmonary arterial pressure (Pap)

●Pulmonary capillary occlusion pressure (PCOP; pulmonary capillary wedge pressure [PCWP])

●Cardiac output (CO)

●Mixed venous oxyhemoglobin saturation (SvO₂)

The PAC can also indirectly measure the following:

Importantly, PAC measurements, particularly in the critically ill patient are frequently inaccurate [1-6]. Reasons for inaccuracies are discussed separately. (See "Pulmonary artery catheterization: Interpretation of hemodynamic values and waveforms in adults", section on 'Sources of error'.)

INDICATIONS — Based upon our clinical experience and that of other experts, pulmonary artery catheters (PACs) are most commonly placed for the evaluation and/or management of patients with the following (table 1) [7-13]:

●Unexplained or unknown volume status in shock

●Severe cardiogenic shock (eg, acute valvular disease, suspected pericardial tamponade)

●Suspected or known pulmonary artery hypertension

●Severe underlying cardiopulmonary disease (eg, congenital heart disease, left-to-right shunt, severe valvular disease, pulmonary hypertension) who are undergoing corrective or other surgery

Large observational studies, small randomized studies, and meta-analyses have confirmed no benefit (survival or days in hospital) from PACs in any medical or surgical population of critically ill patients [7-9,14-29], such that most societal guidelines no longer support their routine use [23,30-33]. As a result, there has been a trend toward decreased routine use of PACs in the United States [34,35]. However, we and other clinicians experienced in the care of severely ill patients, value the information that can be obtained from PAC placement in the select populations listed above. In non-critically ill patients, the insertion of PACs is most commonly performed during the evaluation of patients with suspected or known pulmonary hypertension, the details of which are discussed separately. (See "Clinical features and diagnosis of pulmonary hypertension of unclear etiology in adults", section on 'Right heart catheterization'.)

Despite the debate about the benefits of PA catheters for operative management, an international survey found that practitioners continue to prefer using PA catheters for most cardiac surgery [36]. A large, retrospective analysis of PA catheters in high risk patients undergoing cardiac surgery, however, failed to demonstrate a benefit on morbidity and mortality and, further, was associated with longer length of stay in the intensive care unit (ICU) and increased duration of mechanical ventilation [37]. (See "Anesthesia for cardiac surgery: General principles", section on 'Intravascular cardiac monitors'.)

Unexplained or unknown volume status in shock — Most patients with shock can be classified using a comprehensive diagnostic evaluation that encompasses clinical assessment, laboratory studies, imaging, and bedside ultrasound and/or echocardiography. However, PACs are occasionally inserted in patients with shock in whom a diagnosis has not been achieved with this approach, and/or in patients with unknown intravascular volume status who have been actively resuscitated but remain in shock. In addition, PAC placement in this population can facilitate management decisions regarding fluids and vasopressors as well as evaluate the hemodynamic response to mechanical ventilator changes (table 1). (See "Evaluation of and initial approach to the adult patient with undifferentiated hypotension and shock".)

A widely cited observational study published in 1996 evaluated the effect of PAC placement, performed within the first 24 hours, on the survival of patients admitted to the ICU, most of whom had shock (classified and unclassified) [14]. Patients who underwent PAC had an increased 30-day mortality (odds ratio 1.24) compared to those who did not undergo the procedure. Subgroup analysis did not reveal any group in which PAC was associated with an improved outcome. Despite flaws in the methodological design of this study (eg, influence of vasopressor use, interpreter error) [19-22], additional small randomized studies and meta-analyses have similarly concluded no benefit when PACs are placed (survival or length of hospital stay) [8,14,17,23-25]. Consequently, PACs are no longer used in the routine management of critically ill patients with shock but are reserved for those in whom the acquisition of hemodynamic measurements may help to classify the type of shock (eg, distributive, cardiogenic, hypovolemic, obstructive) or facilitate management decisions [34]. (See "Definition, classification, etiology, and pathophysiology of shock in adults".)

Severe cardiogenic shock — Many experts place PACs in patients who are severely ill with cardiogenic shock. They may also be placed when there is a need to differentiate between cardiogenic and non cardiogenic pulmonary edema (eg, when echocardiography is not helpful). The PAC can also be useful in the diagnosis of pericardial tamponade when clinical evaluation and echocardiography are unhelpful but the suspicion remains. (See "Cardiac tamponade" and "Management of refractory heart failure with reduced ejection fraction", section on 'Use of pulmonary artery catheter'.)

Similar to general medical ICU patients, data supporting the value of PAC in patients with severe heart failure are equally lacking. As an example, one randomized study of 433 patients with heart failure (ESCAPE) reported no difference in mortality (10 versus 9 percent), or the length of hospital stay (8.7 versus 8.3 days) between those who had or did not have a PAC in place [26]. In-hospital adverse events were more common among patients in the PAC group (47 versus 25 percent). Another retrospective study of 1414 patients with cardiogenic shock reported improved in-hospital mortality among patients in whom complete PAC data were obtained compared with those who had no or incomplete PAC data [38].

Patients with suspected or known pulmonary artery hypertension — The PAC is the definitive procedure for the diagnosis and management of patients with pulmonary hypertension, the details of which are discussed separately. (See "Clinical features and diagnosis of pulmonary hypertension of unclear etiology in adults", section on 'Right heart catheterization'.)

Patients undergoing high risk surgery — The PAC is no longer used routinely in high risk patients undergoing elective or emergent surgery [30-32]. However, some experts place PACs perioperatively for select patients considered at high risk for hemodynamic compromise intraoperatively or postoperatively. Examples include those with congenital heart disease, left to right shunt, severe valvular disease, or severe underlying cardiopulmonary disease who are undergoing corrective or other surgery.

Similar to patients who are critically ill, the routine use of PACs perioperatively is of unclear benefit. A large, multicenter trial randomly assigned nearly 2000 high-risk (ASA class III or IV (table 2)) patients ages 60 years or older undergoing cardiac, vascular, or orthopedic surgery to perioperative management with or without a PAC [15]. Patients without a PAC in place were managed on the basis of directly measured central venous pressure and other clinical parameters. There were no significant differences in in-hospital mortality (7.8 versus 7.7 percent) or in one-year mortality (17 versus 16 percent) between the two groups. In addition, a higher rate of pulmonary embolism was reported in the PAC group (0.9 versus 0 percent).

Other — Under rare circumstances, aspirates from a wedged catheter can be used to diagnose lymphangitic carcinomatosis, and may lend support to the diagnosis of fat embolism syndrome [39,40]. While PACS can be useful in this regard, they are rarely placed specifically for this purpose because they are generally insensitive diagnostically. (See "Pulmonary tumor embolism and lymphangitic carcinomatosis in adults: Diagnostic evaluation and management", section on 'Diagnosis' and "Fat embolism syndrome", section on 'Diagnosis'.)

CONTRAINDICATIONS — Absolute contraindications to pulmonary artery catheter (PAC) placement include:

●Infection at the insertion site

●The presence of a right ventricular assist device

●Insertion during cardiopulmonary bypass

●Lack of consent

Relative contraindications to the placement of a PAC include a coagulopathy (international normalized ratio >1.5), thrombocytopenia (platelet count <50,000/microL), electrolyte disturbances (hypo- or hyper-kalemia, -magnesemia, -natremia, -calcemia), and severe acid-base disturbances (eg, pH <7.2 or >7.5). While a bleeding diathesis increases the risk of hemorrhage, electrolyte and acid-base disturbances increase the risk of persistent life-threatening arrhythmias. Correction of these laboratory abnormalities is required prior to inserting the catheter.

The insertion of a PAC in patients with severe pulmonary hypertension and Eisenmenger syndrome is considered high risk (due to the risk of pulmonary artery rupture). Other high risk patients include those with right or left bundle branch block (risk of complete heart block), a defibrillator or pacemaker (risk of displacement), prosthetic or stenotic tricuspid or pulmonary valve (difficult catheter advancement), latex allergy (most catheters have latex), a persistent left superior vena cava (misplacement to the left atrium), and right atrial or ventricular masses (difficult catheter advancement). However, a PAC may be placed for a well-defined indication in these populations usually under fluoroscopic guidance with appropriate resuscitative equipment in place, preferably in a center with expertise in right heart catheter placement. (See "Pulmonary hypertension with congenital heart disease: Clinical manifestations and diagnosis" and "Pulmonary hypertension in adults with congenital heart disease: General management and prognosis", section on 'Management of procedures'.)

MAKING THE DECISION TO PLACE A PULMONARY ARTERY CATHETER — We, and others, agree that in critically ill or perioperative patients, the decision to place a pulmonary artery catheter (PAC) should be based upon a specific question regarding a patient's hemodynamic status or diagnosis that cannot be satisfactorily answered by clinical or noninvasive assessment [41]. If the answer could change management, then placement of the catheter is appropriate. Thus, there is no indication for the routine use of PACs in the intensive care unit or intraoperatively such that clinicians should carefully consider the clinical risks and benefits on a patient-by-patient basis. Importantly, clinicians overseeing catheter placement and collection of hemodynamic data from PACs should be able to assess the quality of the data and reproducibly interpret the results. (See "Pulmonary artery catheterization: Interpretation of hemodynamic values and waveforms in adults".)

Right heart catheterization is frequently performed in patients with clinical and/or echocardiographic suspicion of pulmonary arterial hypertension so that a definitive diagnosis can be obtained. Details regarding patient selection for PAC placement during the evaluation of patients with pulmonary hypertension are discussed separately. (See "Clinical features and diagnosis of pulmonary hypertension of unclear etiology in adults", section on 'Initial diagnostic evaluation (noninvasive testing)'.)

COMPLICATIONS — There are three categories of complications related to the use of pulmonary artery catheters (PACs):

●Those related to insertion of the PAC

●Those related to maintenance and use of the PAC

●Those related to the interpretation of hemodynamic data derived from the PAC

Complications specific to PACs are discussed in the sections below. General complications of central venous cannulation (ie, introducer insertion) including hemorrhage, pneumothorax, and infection (table 3) are discussed separately. (See "Central venous catheters: Overview of complications and prevention in adults" and "Intravascular catheter-related infection: Epidemiology, pathogenesis, and microbiology".)

Insertion — Common complications of insertion include the occurrence of atrial and/or ventricular arrhythmias. Less commonly, catheter misplacement or knotting can occur. Perforation of a cardiac chamber and rupture of a cardiac valve or the pulmonary artery are rare complications that can be catastrophic. Many of these complications can be minimized, but not eliminated, by paying careful attention to the pressure waveforms during insertion and by ensuring that the balloon at the tip of the catheter is fully inflated, so that the catheter tip does not protrude beyond the balloon (figure 1). (See "Pulmonary artery catheters: Insertion technique in adults".)

Arrhythmias — Atrial and ventricular arrhythmias commonly occur during passage of the catheter through the cardiac chambers. They are easily identified on the bedside monitor. Most of these rhythm disturbances are self-limiting [42]. However, potentially serious rhythm disorders can be seen:

●Sustained ventricular arrhythmias (>30 consecutive ventricular contractions) occur in up to 3 percent of patients [42]. However, because it is an uncommon occurrence, prophylactic use of intravenous lidocaine is not recommended nor is it of proven benefit prior to PAC placement [43-45]. When such severe rhythm disturbances are seen and not eliminated by advancement into the pulmonary artery or balloon deflation and catheter withdrawal, they should be treated according to the usual ACLS protocol. Importantly, significant ventricular arrhythmias may be mitigated by correcting electrolyte and acid-base disturbances prior to insertion. (See "Advanced cardiac life support (ACLS) in adults" and 'Contraindications' above.)

●Right bundle branch block develops during approximately 5 percent of catheter insertions, thereby placing patients with preexisting left bundle branch block (LBBB) at risk for complete heart block [46,47]. Although the risk is low and complete heart block is usually transient, catheter insertion should not be undertaken in patients with LBBB without the ability to institute immediate cardiac pacing [48]. This can be accomplished by placement of a transthoracic or transvenous pacer, or by using a PAC with an intraluminal pacing electrode. In many centers, pulmonary artery catheterization in the setting of a LBBB is performed under fluoroscopic guidance to expedite passage of the catheter tip through the right ventricle and decrease the risk of inducing complete heart block.

Misplacement — Difficulty placing the PAC can occur during insertion due to looping of the catheter inside the right atrium or right ventricle as well as due to anatomic vascular anomalies (eg, atrial septal defect). This can generally be managed with positional maneuvers and/or fluoroscopic guidance, which are discussed separately. (See "Pulmonary artery catheters: Insertion technique in adults", section on 'Advancement through cardiac chambers'.)

Knotting — Knotting of the PAC inside the cardiac chambers should be suspected when the PAC does not reach its intended target chamber or pulmonary artery at the expected distance during advancement or when resistance is met during catheter withdrawal. It may also be recognized on chest radiograph. When knotting is strongly suspected, the catheter should not be withdrawn and imaging, usually chest radiograph but occasionally computed tomography, should be obtained for diagnosis. The knot may be relieved by injecting 10 to 20 mL of cold sterile saline or of iced solution prepared for cardiac output measurement through the distal lumen ("stiffens" the tip); the PAC can then be removed transvenously. However, some patients require placement of a guidewire in a cardiac catheter laboratory under fluoroscopic guidance venotomy [49], or even surgical extraction [50].

Myocardial, vessel, or valve rupture — Endocardial injury and valve inflammation or damage can occur during PAC insertion which may be asymptomatic and transient. Acute myocardial, vessel, or valve rupture should be suspected if sudden respiratory distress and shock develop during insertion, the clinical presentation of which is discussed separately. (See "Etiology, clinical features, and evaluation of tricuspid regurgitation" and "Acute myocardial infarction: Mechanical complications", section on 'Clinical presentation' and 'Use and maintenance' below.)

Use and maintenance — Complications associated with use and maintenance of the catheter include pulmonary artery perforation, pulmonary infarction, thromboembolic events, and infection [20,21,43]. These complications can often be avoided by careful attention to catheter position and strict adherence to sterile technique when using the catheter. Thus, they are more likely to be encountered when PACs are left in for a prolonged period of time (eg, critically ill patients) and less likely in those who have catheters placed for a few hours only (eg, catheters placed for the diagnosis of pulmonary hypertension).

Pulmonary artery perforation — The most feared complication of PAC placement is pulmonary artery perforation, which has a mortality that ranges from 30 to 70 percent [51]. This complication may occur during insertion, but often occurs later when the catheter has inadvertently migrated into a distal position in the pulmonary artery. Rupture can be due to perforation by the catheter tip or balloon overinflation. It should be suspected in a patient with a PAC who presents with brisk hemoptysis. While some cases are not life-threatening and can be managed conservatively, hemoptysis is often massive and usually requires emergent thoracotomy for management. In addition, when pulmonary artery rupture is suspected or confirmed, the catheter should be removed. Details regarding the management of hemoptysis are discussed separately. (See "Evaluation of nonlife-threatening hemoptysis in adults" and "Evaluation and management of life-threatening hemoptysis".)

Self-limiting pulmonary hemorrhage following perforation can result in the formation of a pulmonary artery pseudoaneurysm, which is at risk of subsequent hemorrhage that can be fatal [43,52-55]. Although a pseudoaneurysm may be suspected by visualizing as a nodule or infiltrate on chest radiography at the PAC tip, the diagnosis usually is made on contrast-enhanced computed tomography of the thorax (image 1) [52]. Pulmonary angiography can be used to confirm the diagnosis (image 2), and may permit direct coil embolization of the lesion [52,54,55].

Risk factors for pulmonary artery rupture include pulmonary hypertension, advanced age, mitral valve disease, hypothermia, and anticoagulant therapy [56]. For patients at increased risk, it may be advisable to use the pulmonary artery diastolic pressure to approximate the pulmonary artery occlusion pressure, if possible, rather than repeatedly inflate the balloon. Conditions in which the pulmonary artery diastolic pressure does not approximate the wedge pressure are discussed separately. (See "Pulmonary artery catheterization: Interpretation of hemodynamic values and waveforms in adults".)

Pulmonary infarction — Migration of the catheter tip into the distal branches of the pulmonary artery or a balloon left inflated in the wedge position for an extended period of time may result in pulmonary infarction. When small, these infarcts may be asymptomatic and diagnosed only by chest computed tomography. Larger infarcts may present with classic symptoms of pleuritic chest pain, hemoptysis, and hypoxemia. If PAC-induced infarction occurs, the catheter should be removed and the patient anticoagulated, provided there is no contraindication. (See "Clinical presentation, evaluation, and diagnosis of the nonpregnant adult with suspected acute pulmonary embolism" and "Treatment, prognosis, and follow-up of acute pulmonary embolism in adults".)

Thromboembolism — Thromboembolic events may occur with catheters acting as a nidus for thrombus formation. Mural thrombi induced by inflammation or infection of an endocardial or vessel wall are present in up to 30 percent of patients [57]. These clots can propagate (and cause vessel occlusion), embolize, or act as a nidus for infection. This complication is now seen less frequently because of the common use of heparin-bonded catheters [58]. High-risk groups include those with hypercoagulable states, traumatic insertions, or prolonged need for monitoring [56]. While use of heparin prophylaxis may theoretically decrease this risk, the impact of either prophylactic or full dose anticoagulation in this population is unknown.

Infection — Endocarditis, especially of the pulmonic valve, occurs in less than 2 percent of cases [59]. Catheter-related infections, with possible bloodstream infection can also occur. (See "Intravascular catheter-related infection: Epidemiology, pathogenesis, and microbiology" and "Intravascular non-hemodialysis catheter-related infection: Clinical manifestations and diagnosis".)

Other — Other complications seen with PACs include:

●Sterile cardiac valve vegetations are seen in up to 30 percent of cases when catheters are left in place for prolonged periods [14]

●Venous air embolism, caused by open infusion ports or catheter hubs, can lead to entrainment of room air through the monitoring system and into the venous circulation (see "Air embolism")

Data misinterpretation — PAC-derived hemodynamic data can be inaccurate or misinterpreted for several reasons [20-22,60,61]:

●Improperly calibrated or leveled pressure monitors (see "Pulmonary artery catheters: Insertion technique in adults", section on 'Preparing the catheter')

●Transduction of airway pressures under non-zone 3 conditions (figure 2) (see "Pulmonary artery catheterization: Interpretation of hemodynamic values and waveforms in adults", section on 'Lung zone misplacement')

●Over- or under-estimation of the pulmonary capillary wedge pressure due to under- or over-inflation of the balloon, respectively, which in turn is often due to misplacement or migration of the catheter (see "Pulmonary artery catheters: Insertion technique in adults", section on 'Final wedge position')

●Interobserver variability in the interpretation of hemodynamic data, even among physicians experienced in waveform interpretation

For these reasons, misinterpretation of data is generally accepted to be common and can potentially lead to mismanagement of the patient. (See "Pulmonary artery catheterization: Interpretation of hemodynamic values and waveforms in adults".)

In light of the data suggesting decreased use of PACs, data misinterpretation can reasonably be expected to increase in frequency as clinicians lose valuable experience using the catheter and interpreting its data.

FUTURE DIRECTIONS — Several less invasive approaches to assessing hemodynamic data in critically ill patients are being increasingly used by some intensivists. These include:

●Dynamic indices – Dynamic indices have been studied as a potential target to guide fluid management in patients with shock, in particular, septic shock. The principle of dynamic measurements is that they all measure a specific hemodynamic index in response to a stimulus (eg, fluid bolus, respiration) as a mechanism of assessing a patient's intravascular volume.

While central venous pressure (CVP), mean arterial pressure (MAP), mixed venous oxyhemoglobin saturation (SvO₂) and pulmonary artery wedge pressure are considered "static" hemodynamic measures, respiratory changes in the vena caval diameter, radial artery pulse pressure, aortic blood flow peak velocity, and brachial artery blood flow velocity are considered "dynamic" hemodynamic measures [62,63]. There is increasing evidence that dynamic measures are more accurate predictors of fluid responsiveness than static measures, as long as the patients are in sinus rhythm and passively ventilated with a sufficient tidal volume [64-66]. For actively breathing patients or those with irregular cardiac rhythms, an increase in the cardiac output in response to a passive leg-raising maneuver (measured by echocardiography or arterial pulse waveform analysis) is a sensitive and specific predictor of fluid responsiveness [67]. (See "Novel tools for hemodynamic monitoring in critically ill patients with shock".)

●Transesophageal echocardiography – Transesophageal echocardiography with Doppler is used by some anesthesiology experts perioperatively to assess cardiac preload and cardiac output [68,69]. Similarly, bedside echocardiography and thoracic ultrasonography are being increasingly used to assess volume status of patients in the intensive care unit. (See "Hemodynamics derived from transesophageal echocardiography" and "Indications for bedside ultrasonography in the critically ill adult patient".)

●Electrical bioimpedance – Electrical bioimpedance systems measure subtle changes in electrical conduction through the body over the course of the cardiac cycle to determine stroke volume using pulse contour analysis algorithms [70-72].

●Lithium dilution – Lithium dilution techniques analyze changes in lithium concentration over time following a bolus injection to estimate cardiac output [73,74].

●Inert gas rebreathing – Inert gas rebreathing systems use low-concentration soluble and insoluble inert gases to calculate the cardiac output by the Fick principle [74]

Although some systems based on these techniques have shown good correlation with cardiac output measured via PAC thermodilution [71,74], clinical experience is limited and further study will be needed before any of these approaches can be recommended [75,76].

SUMMARY AND RECOMMENDATIONS

●Pulmonary artery catheters (PACs) obtain direct measurements of central venous, right-sided intracardiac, pulmonary artery, and pulmonary artery occlusion pressures. In addition, they can estimate cardiac output, systemic and pulmonary vascular resistance as well as mixed venous oxyhemoglobin saturation, oxygen delivery, and oxygen uptake. (See 'Physiologic measurements' above.)

●In critically ill patients, there is no benefit to the routine insertion of PACs. However, in select populations, we and other experts believe that valuable information can be obtained from PACs. The most common indications for PAC placement are the evaluation and/or management of patients with unexplained or unknown volume status in shock, severe cardiogenic shock, and suspected or known pulmonary arterial hypertension (PAH) (table 1). Some experts also place PACs perioperatively in patients with congenital heart disease, severe valvular disease, left-to-right shunt, and severe underlying cardiopulmonary disease. (See 'Indications' above.)

●Absolute contraindications to PAC placement include infection at the insertion site, right ventricular assist device, and insertion during cardiopulmonary bypass, as well as lack of consent. Relative contraindications to the placement of a PAC include the presence of a coagulopathy (international normalized ratio >1.5), thrombocytopenia (platelet count <50,000/microL), electrolyte disturbances (hypo- or hyper-kalemia, -magnesemia, -natremia, -calcemia), and severe acid-base disturbances (eg, pH <7.2 or >7.5). (See 'Contraindications' above.)

●The decision to place a PAC in critically ill or perioperative patients should be made on a case-by-case basis after carefully weighing the potential benefits against the risks for the individual patient. The decision to perform right heart catheterization in patients suspected to have PAH is dependent upon a high suspicion for PAH following a comprehensive diagnostic evaluation. (See 'Making the decision to place a pulmonary artery catheter' above and "Clinical features and diagnosis of pulmonary hypertension of unclear etiology in adults", section on 'Right heart catheterization'.)

●Common complications of PAC insertion include the occurrence of atrial and/or ventricular arrhythmias. Less commonly, catheter misplacement or knotting can occur. Perforation of a cardiac chamber and rupture of a cardiac valve or the pulmonary artery are rare complications that can be catastrophic. Complications of catheter use include pulmonary artery rupture, pulmonary infarction, thromboembolic events, infection, and data misinterpretation. (See 'Complications' above.)

●Several less invasive approaches to assessing hemodynamic data in critically ill patients have been proposed. These include dynamic indices and transesophageal echocardiography, as well as electrical bioimpedance, lithium dilution, and inert gas rebreathing systems. (See 'Future directions' above.)