Pulmonary Artery Pressure Monitoring

Elizabeth J. Bridges is an Assistant Professor, Biobehavioral Nursing and Health Systems, University of Washington School of Nursing, 1959 NE Pacific St, Seattle, WA 98195 (e-mail: [email protected]). been completed to evaluate the effects of the use of PA catheters as a part of care. The metaanalysis of 13 randomized clinical trials found no difference in mortality or length of stay between groups of patients who were randomized to PA catheter versus no PA catheter groups. The Evaluation Study of Congestive Heart Failure and Pulmonary Artery Catheterization Effectiveness (ESCAPE) evaluated the efficacy and safety of PA catheter use in patients with acute heart failure. In this study, endpoints for resuscitation were specified, but there were no standardized recommendations for the use of diuretics or inotropic agents. The study found that the use of a PA catheter did not increase or decrease the mortality or VOLUME 17 • NUMBER 3 • JULY–SEPTEMBER 2006 PA PRESSURE MONITORING 287 length of stay for acute heart failure patients; however, the group that received PA catheter adjusted therapy had a greater improvement in quality of life and a trend toward improved exercise capacity. The Pulmonary Artery Catheter in the Management of ICU Patients (PAC-Man) study evaluated PA catheter versus use of an alternative method for cardiac output monitoring (eg, transesophageal Doppler) in critically ill patients with ARDS, heart failure, and multiorgan dysfunction. No specific treatment guidelines or endpoints were used. There was no significant difference between groups in ICU or 28-day mortality; although, in the PA catheter group, 80% of patients had a change in treatment made within 2 hours of insertion. The Sepsis Occurrence in Acutely Ill Patients (SOAP) study was an observational study that evaluated the possible association between PA catheter use and outcome. Although patients with PA catheters had a higher mortality rate, when confounding factors such as acuity, age, organ dysfunction, and comorbidities were controlled for, the use of a PA catheter was not independently associated with a higher risk of 60-day mortality. Finally, a study of the use of PA catheters in patients with shock, ARDS, or both found no difference in 14 or 28-day mortality in patients treated with routine (nonstandardized) treatment. A critical point when reviewing these studies is that the insertion of a PA catheter and simply monitoring hemodynamic indices does not improve outcomes. In addition, it may be insufficient to specify endpoints as goals to be met and not specify the most effective treatment regimen to reach these endpoints. Hemodynamic indices (to include perfusion indices) should be used as a part of an evidence-based treatment plan aimed at optimizing tissue perfusion before organ dysfunction occurs, as exemplified by the improved outcomes associated with goal-directed therapy for patients with septic shock, high-risk surgical patients, and postcardiac surgery patients. The challenge remains to identify which hemodynamic indices and types of monitoring devices (eg, PA catheter, noninvasive pressure and cardiac output monitors, perfusion indices) improve outcomes, to identify specifically which patient populations (ie, patient type, etiology of shock, severity of illness, and timing of interventions) will benefit most from the integration of hemodynamic data into their care, and to develop population specific protocols. Although there are clinical conditions in which the integration of hemodynamic data may improve diagnostic accuracy and integration of hemodynamic data into a plan of care will improve outcomes, there are 2 general areas that limit the utility of PA pressure monitoring. First, critical care clinicians (nurse and physicians) may incorrectly gather and interpret the data. Several excellent resources to improve the knowledge and ability to interpret and use PA pressure data are the Pulmonary Artery Catheter Education Program (PACEP), which is a series of Web-based training modules that cover clinical and technical aspects of care and waveform interpretation and AACN’s Practice Alert: Pulmonary Artery Pressure Monitoring. Second, while the pulmonary artery occlusion pressure (PAOP) provides useful information regarding hydrostatic pressure and the risk for pulmonary edema, static hemodynamic indices (RAP, PAOP) are not sensitive or specific indicators of preload or fluid responsiveness. The remaining sections of this article focus on: (1) a review of factors that affect the reliability and accuracy of hemodynamic indices (ie, zeroing, referencing, dynamic response characteristics, and filter frequency), (2) how to interpret the hemodynamic data within the context of complex care situations (eg, proning, marked respiratory variation, cardiogenic versus noncardiogenic pulmonary edema), and (3) functional hemodynamic measures, which are an alternative to static hemodynamic measurements. Factors That Affect the Reliability and Accuracy of Pulmonary Artery/Right Atrial Pressures Zeroing Zeroing of the pressure monitoring system is performed by opening the system to air to establish atmosphere as zero. The current transducers have minimal zero drift and do not require routine rezeroing. Of note, rezeroing is not the same as rereferencing, which is required with any change in the patient’s position relative to the transducer. A recommendation based on older transducer technology was that the pressure system should be rezeroed with changes in barometric pressure, such as those that occur with a change in the weather or ascent to altitude during BRIDGES AACN Advanced Cri t ical Care 288 aeromedical transport. The current pressure transducers are vented to air and do not require routine rezeroing with a change in barometric pressure. Referencing Referencing, which is performed to correct for the changes in hydrostatic pressure above and below the heart, is accomplished by placing the air-fluid interface (stopcock) of the catheter system at the level of the heart to negate the weight effect of the catheter tubing. Numerous texts identify the midaxillary line as the reference point; however, in the supine position use of the midaxillary line as the reference rather than one-half the anterior/posterior diameter of the chest, may result in a measurement error of 6 mm Hg in patients with varied chest wall configurations. Similarly, the use of angle specific references is also required for the lateral position (see Figure 4). Dynamic Response and Filtering The dynamic response characteristics of the system affect the ability of the pressure monitoring system to faithfully reproduce a pressure waveform. A method to evaluate and optimize a pressure monitoring system has been previously described. One key point in optimizing a pressure system is that air bubbles affect the dynamic response characteristics of any pressure monitoring system. Thus, measures must be taken during the set-up and maintenance of the pressure system to remove air bubbles. The “rocket flush,” which is a 10 mL manual rapid flush of the system, starting at the proximal stopcock, is an additional step during line preparation to remove air from the system. The performance of a rocket flush during line preparation significantly improves the dynamic response characteristics of pressure monitoring systems (Figure 1). The “rocket flush” should never be performed when the system is attached to a patient because of the risk of retrograde air embolization. A validated algorithm that decreases air bubble formation and optimizes the dynamic response characteristics of the pressure system is provided (Table 1). Adjusting the filter frequency limits on the monitor is often incorrectly attempted when the pressure system is underdamped or when there is catheter whip. To understand the function of the filter, it is important to understand how a waveform is created by the pressure monitoring system. The oscillations caused by the pressure wave striking the fluid in the catheter causes distortion of the diaphragm in the transducer and the creation of an electrical signal. The electrical signal from the transducer is sent to the monitor where it is amplified, filtered, and converted to the waveform and digital output observed on the monitor. The waveform that is observed on the monitor is a summation of a series of sine waves or harmonics (Figure 2). For a patient with a heart rate of 60 beats per minute, the fundamental frequency (first harmonic) is equal to the pulse rate and occurs at one cycle/second (1 Hz). The second harmonic occurs at 2 Hz, the third harmonic at 3 Hz, etc. For a patient with a heart rate of 120 beats per minute, the first harmonic occurs at 2 Hz, the second harmonic at 4 Hz, the third harmonic at 6 Hz, etc. The important physiological information is contained in the first 6 to 10 harmonics (12 to 20 Hz). Thus, the bandwidth of the filter on the monitor should be set to allow for up to 12 to 20 Hz for a patient with a heart rate of 120 beats per minute. Figure 1: Effect of a “rocket flush” on the dynamic response characteristics of a pressure monitoring system. A, Control: pressure system with a VAMP (PXVMP160; Edwards LifeSciences, Irvine, Calif.) after standard set-up. Fn, 9 Hz; amplitude ratio, 0.3; dynamic response, underdamped. B, Same system after a 10 mL “rocket flush.” Fn, 13 Hz; amplitude ratio, 0.4; dynamic response, adequate. VOLUME 17 • NUMBER 3 • JULY–SEPTEMBER 2006 PA PRESSURE MONITORING Decreasing the filter frequency below 12 Hz may cause the waveform to appear cleaner, but important physiological waveform information is being filtered; thus, a true waveform is not being observed. Rather than adjusting the filter to clean up the waveform, attention should be paid to optimizing the dynamic response characteristics of the system. Pulmonary Factors There are numerous pulmonary factors that add challenges to accurately and reliably obtaining PA pressure measurements: (1) the use of digital versus analog data, (2) how to determine if the PA catheter tip is in the correct lung zone, (3) how to interpret the pressures when there is marked respiratory variation, and (4) interpretation of invasive pressures when there is increased pleural pressure. Analog Versus Digital Data An assumption of invasive pressure monitoring is that the measured pressure (eg, pulmonary artery end diastolic pressure [PAEDP]/ Table 1: Invasive Pressure Monitoring Line Preparation