The cardiac surgeon states that he would like a pulmonary artery catheter (PAC) instead of the central venous line. He would like to know the pulmonary artery pressure and monitor it during the placement of the VAD and in the immediate postoperative period. A wire is placed in sterile fashion through the most distal lumen of the CVL. The CVL is then removed and the PAC is placed via a sheath introducer and the catheter is attempted to be floated in the pulmonary artery. The patient develops ectopy and subsequently ventricular tachycardia that does not improve with removal of the catheter from the ventricle. The patient is immediately cardioverted and no further attempts are made to float the Swan–Ganz at this time.

Pulmonary artery catheters (PAC) have been used in cardiac surgery for measuring cardiac output (CO), mixed venous saturation, direct pulmonary artery pressures, and pulmonary capillary wedge pressure. They were routinely used until a number of studies including a Cochrane review and the ESCAPE trial questioned their effectiveness in the ICU and in the operating room (Pulmonary Artery Catheter Consensus conference: consensus statement 1997; Binanay et al. 2005; Rajaram et al. 2013). These studies, among many others, have resulted in a marked reduction in the number of PACs placed, though many institutions still place them routinely (Marik 2013b). The placement of a PAC in children is very uncommon and is often complicated by sheath size and intracardiac shunting that make the readings unreliable or inaccurate. Though now almost 20 years old, the Pulmonary Artery Consensus Conference recommended the use of PA catheters in children suffering from shock refractory to fluids and vasopressors, pulmonary hypertension, and acute lung injury when attempting to decipher cardiogenic from non-cardiogenic causes. A recent review supported this statement and discussed the lack of evidence for their use (Perkin and Anas 2011). There is much more variability in adult patients, but a major meta-analysis looked at 13 randomized studies in ICU, surgical, and cardiac settings showed no significant improvement in any major outcome associated with PAC usage (Shah et al. 2005). In a recent study, the benefits of PACs were discussed in very specific cases including acute heart failure requiring inotropes (Sotomi et al. 2014). Many experts also believe that there are patient-specific situations where placement of a PAC may help management where no guidelines are in place (Kahwash et al. 2011).

When placed, a PAC can give information that cannot be accurately obtained from any other monitor system. CO monitoring can be completed via two means, thermodilution and oxygen consumption. Thermodilution is usually favored in adult patients due to its ease and immediate value obtained without practitioner calculation. In this method, a known volume at a known temperature is injected into the patient via the catheter itself. The thermistor then reads the temperature downstream, and the CO is calculated, dependent on the temperature change and using the patient’s body surface area (Ganz and Swan 1972). The thermistor placed on the catheter requires a larger sheath and, therefore, is often difficult to place in children and infants. Thermodilution is also not accurate when there is intracardiac shunting as seen in many congenital heart patients (Freed and Keane 1978). The Fick method of oxygen saturation sampling is instead routinely used. The Fick method requires the sampling of oxygen saturation at the pulmonary artery (venous oxygenation) and pulmonary vein (arterial oxygenation) with an estimated oxygen consumption based on the size of the patient to calculate CO (Rutledge et al. 2010). The peripheral artery oxygenation is often used as a surrogate for the pulmonary vein, but this assumes no significant intracardiac shunting:
or
PACs can directly obtain the pressure of the pulmonary artery in adults and children. This is often beneficial in patients with pulmonary hypertension though the risk in obtaining them especially in children is high (Carmosino et al. 2007). However, there currently are no other means to obtain accurate pressure measurements across the capillary bed, making placement a necessity. The PA catheter also has a balloon at the end and can be wedged in a peripheral pulmonary artery allowing for an indirect measurement of left atrial (LA) pressure. This can help decipher cardiac vs. noncardiac lung injury.

The placement of a PA catheter is generally safe but includes all the risk of IJV placement and, additionally, the increased risk of placing a large catheter from the right atrium across the tricuspid valve into the right ventricle and then into the pulmonary artery. Care should be taken not to cannulate or dilate the carotid artery as severe injury could occur due to the large sheath and catheter size. The placement of the catheter can cause conduction problems, arrhythmias, and injury to the valves, and there are case reports of it knotting requiring surgical removal (Graybar et al. 1983; Perkin and Anas 2011). The most concerning complication is pulmonary artery hemorrhage that may occur when obtaining a wedge pressure. The hemorrhage is often difficult to stop and is a mortality risk (Hannan et al. 1984). A PA catheter may be placed directly by the surgeon in the operating room if that information is considered important for intra- and postoperative management. Depending upon the catheter type chosen, it is possible to also make use of a continuous mixed venous oximetry catheter to be utilized to help guide postoperative inotropic, transfusion, and volume therapy.

PACs can give information that is difficult if not impossible to achieve through any other monitoring system. They are not without their risks, however, and these risks frequently outweigh any benefit in routine cardiac cases. There are certain patients and medical situations where a PAC is not only warranted but potentially required and great care should be used in its placement and interpretation.

The PAC is floated right before bypass initiation with assistance by the surgeon guiding it manually to minimize the risk of arrhythmias during placement. The wedge pressure is 22, with PA pressures of 54/32 mean of 41 with a mean systemic pressure of 62. The wedge pressure is 8 with a mean PA pressure of 30 compared to the systolic mean of 58 after VAD initiation (Fig. 7.2).

Minimally invasive cardiac output monitors are defined as any device placed that can measure cardiac output without the placement of a PAC. The benefits of CO monitoring without the risks of PAC placement are very appealing. There are multiple devices that have been developed with a great deal of variability in their science and accuracy. Many of these devices require some invasive catheters including arterial, CVL, or both. We will discuss only a few on the many types of minimally invasive monitors in this chapter.

The most commonly used monitors evaluate pulse contour analysis and the systolic upstroke of an arterial line to obtain stroke volume. These monitors must be calibrated, and the two most common use lithium or ice cold water to obtain a thermodilution baseline from CVL or peripheral IV¹ to arterial line. An algorithm is used, while the arterial waveform is continuously monitored, and the CO is displayed. These monitors are inaccurate when there is any dampening or resonance of the arterial signal and are also inaccurate with an aortic balloon pump, arrhythmias, or aortic insufficiency (Monnet et al. 2004; Hofer et al. 2007; Richard et al. 2011; Monnet and Teboul 2015). There is also limited number of studies in the cardiac operating room, and only two studies involve pediatric cardiac patients (Mahajan et al. 2003; Sander et al. 2005, 2006; Fakler et al. 2007; Phan et al. 2011; Broch et al. 2015). The majority of these studies showed relative inaccuracies of the monitors compared to the gold standard methods (Fick or thermodilution). The risk factor of these catheters is minimal and is the same as the risk of the invasive lines needed for them.

The ultrasound technique of noninvasive cardiac monitoring is a by-product of the development of routine transesophageal echocardiography (TEE) used in cardiac and other major surgical cases. TEE is able to estimate CO by obtaining the instantaneous blood flow through a specific cross-sectional diameter of the descending aorta multiplied by the heart rate. The probes used in the minimally invasive technique are much smaller and portable compared to the TEE probes. Each has a different method in obtaining the flow and diameter, and general validation to the gold standard has been poor in most studies (Valtier et al. 1998; Chand et al. 2006; Chatti et al. 2009; Phan et al. 2011).

There is some benefit however in following trends despite the lack of absolute accuracy when compared to thermodilution. The major concerns are the toleration and difficulty with precise placement with the intraesophageal type. The extrathoracic version uses nomograms, which may further decrease accuracy. Both types use the assumption that blood flow is the same in the carotids and the descending aorta, which is not necessarily the case in sick patients (Marik 2013a). The studies in children are extremely poor with no studies including patients with shunts or significant congenital heart disease (Wongsirimetheekul et al. 2014; Beltramo et al. 2016). The risk factors for placement of these devices are minimal for the intraesophageal (similar to OG tube placement) and virtually nonexistent for the extrathoracic version.

Bioimpedance cardiac monitors are the least invasive of the devices discussed in this chapter. Electrodes are placed on a patient that both give and receive electrical signal. The primary component of variable impedance (what is altering the signal) is blood flow in the aorta. An algorithm is used to compute cardiac output from this change in impedance. These monitoring devices would seem to have the poorest correlation with the gold standard of thermodilution (Marik et al. 1997; Critchley et al. 2000; Spiess et al. 2001; Sageman et al. 2002; Gujjar et al. 2008). Pediatric studies have used these devices clinically, showing CO changes with interventions and correlation with TEE, but there was poor correlation with thermodilution (Schubert et al. 2008; Cote et al. 2015). The placement of these devices involves simple electrodes and is without any significant risk.

Minimally invasive cardiac output monitors are being used in many patients with variable diseases and ages. Their accuracy and validity are questioned when compared to the gold standard, but new algorithms have improved some of these concerns. They have yet to be considered part of the routine management for pediatric or adult cardiac patients.

At this time, a monitor that is without flaws and that will diagnose the hemodynamic variable that it is monitoring has not been developed. It is imperative that the practitioner understand how a specific monitor works and its limitations and the risks with its placement before deciding on its use. The interpretation of the monitor and the practitioners’ response will always be the most important component in patient management (Tables 7.3 and 7.4).