The care of obstetric patients requiring hemodynamic monitoring can be quite complex. It requires a basic understanding of pregnancy physiology, monitoring equipment, and applications of information gathered. In this chapter, we will review invasive and noninvasive hemodynamic principles provide the reader with a functional understanding of circulatory monitoring and ways to practically apply this understanding to normal and pathologic conditions in the gravid patient.

The interpretation of data gathered with hemodynamic monitoring is most useful when applied according to disease state and used to identify ominous changes before they result in clinical deterioration (Table 1-1). Responses to therapy can be closely followed with the help of hemodynamic monitoring using invasive and noninvasive techniques.

Functional Hemodynamics

Invasive Monitoring

J.C. Swan and W. Ganz first introduced this technology in 1970.¹ The pulmonary artery catheter (PAC) is guided into the superior vena cava (SVC) by way of the internal jugular or subclavian vein, then into the RA, with blood flow it “floats” into the RV, past the pulmonary valve into the pulmonary artery and the inflated balloon tip will end up positioned or “wedged” into a branch of the pulmonary artery. Measurements should be taken at end-expiration (Fig. 1-1).

Once “wedged”, the PAC waveform has a specific configuration, each portion relates to an action in the left atrial chamber. The “a-wave” correlates to atrial contraction, one will notice a rise in pressure on the waveform. Next is the “x-wave” which signifies atrial diastole and relaxation of the chamber in preparation for filling, one will notice a drop in pressure on the waveform. The next rise in pressure corresponds to passive left atrial filling and this is the “v-wave”. The “y-wave” denotes atrial emptying (Figs. 1-2 and 1-3).

CO is often measured with a PAC using thermodilution. A small volume of liquid with a lower temperature than the blood is injected into the PAC where a thermistor measures the change in temperature as the blood flows through the heart chambers. These temperature change measurements can be plotted on a graph and the area under the curve (AUC) is used to estimate the cardiac output (Fig. 1-4) (Table 1-2).¹

Over the past two decades, the utility of the PAC in guiding the management of critically ill patients has been under scrutiny. Keeping in mind, outside of pregnancy, several multicenter studies have been unable to demonstrate mortality benefit with the use of a PAC at 30 days, 6 months, and 1 year in ICU and high-risk perioperative patients.^1-3 Clinicians overall have become less adept at interpreting data from PAC measurements as they have been used less frequently over time. Aside from that, PAC placement is not without risk. Arrhythmias, iatrogenic pneumothorax, vascular rupture and thrombosis, valve injury, and central line blood stream infection can occur. Although the prevalence of these risks in recent literature range from 0.1% to 1.3%,^4,5 the consequence of a complication can be dire.

Evidence for PAC use in pregnancy is scarce in modern literature. In obstetrics, the primary indication for invasive hemodynamic monitoring has been in women with severe preeclampsia complicated by acute kidney injury (AKI) and oliguria, as well as those with pulmonary edema where PAC has been shown to facilitate management.⁶ Other studies and reports have shown it to be helpful in the management of women with severe valve disease, pulmonary hypertension, cardiomyopathy, septic shock, and renal disease.⁷

The above said, information gathered from PAC measurements effectively provides a means to estimate the impact of the parameters that effect how well the heart pumps and the response to treatment. These parameters include: venous return/preload, peripheral resistance/afterload, heart rate, and contractility. Functionally, if one wants to determine the effect or adequacy of a volume load in patients who are fluid avid, you want to see an increase in blood pressure with a corresponding decrease in heart rate, as well as an increase in central venous pressure (CVP), pulmonary artery occlusion/wedge pressure (PAOP), stroke volume variation (SVV), and ideally cardiac output (CO).⁸ Those patients who respond to volume can be safely given additional boluses with a low risk of developing pulmonary edema until adequate resuscitation is determined. Keep in mind that no one variable alone is an adequate measure of volume response when looked at independently, as noted in the FACCT study.⁹ It is the combination of measurements and trends that are most helpful in determining volume response.

There are specially made PACs that can measure oxygen saturation continuously and accurately estimate the mixed venous oxygen saturation (SVO₂) at the level of the distal pulmonary artery. The SVO₂ represents the amount of oxygen left in the circulation after passing through the tissues and measures the balance between oxygen consumption and delivery. The normal value for SVO₂ is 70% to 75%.¹⁰ It makes sense that SVO₂ is decreased in conditions that increase consumption, for example: exercise, shivering, severe anemia, hypoxemia, and decreased cardiac output. A very low SVO₂ (ie, 30%-40%) is seen with tissue ischemia.¹⁰

The CVP is an estimate of right ventricular preload. It can be elevated in cases of pulmonary edema, positive pressure ventilation, right ventricular failure (more commonly as a sequlae of left ventricular failure), pulmonary hypertension, severe pulmonary stenosis and tamponade. One CVP measurement alone is not helpful, the trend in measurement allows one to monitor response to therapy (ie, volume or diuresis) is most useful. One functional application of this is in patients who are receiving positive pressure ventilation. If the patient is intravascularly depleted, as compared to those who are euvolemic, you will see a decrease in cardiac output with application of positive pressure (inspiration phase of mechanical ventilation) because the vena cava get compressed in the thorax and this decreases venous return. The CVP in this circumstance may measure a little higher than expected because of increased intrathoracic pressure from positive pressure ventilation but the patient is still low in intravascular volume. In patients with normal hearts who are spontaneously breathing, the negative (decreased) intrathoracic pressure with inhalation “pulls blood into the heart” and increases venous return. The CVP typically decreases by 1 mm Hg during inhalation in spontaneously breathing patients.⁸ However, in spontaneous ventilation, it can decrease by more with intravascular depletion.

Measurements of the PAOP are used to indirectly estimate left atrial pressure or the pressure in the pulmonary veins. Following the trends in PAOP can be helpful in estimating pulmonary venous resistance and left atrial preload.¹ PAOP measurements are helpful in distinguishing cardiogenic (with both high PAOP and CVP) from noncardiogenic (low or normal PAOP and high CVP) pulmonary edema.¹

In pathologic conditions that involve pulmonary capillary wall damage or increased permeability (severe preeclampsia, sepsis, ARDS, hemorrhagic shock in extremis), the CVP and PAOP can be anticipated not to correlate well.¹¹ Interestingly enough, disparate values between the CVP and PAOP have been demonstrated in gravid patients without left heart failure and a structurally normal heart.¹² Because of this, CVP readings in pregnant patients cannot be used to make any assumptions about function of the left side of the heart and are most reliable at the extremes of values (really low or really high) (Table 1-3).¹²

Noninvasive Monitoring

Belfort et al¹³ was able to demonstrate a reasonable correlation between invasive and noninvasive monitoring by 2D echocardiogram in 11 critically ill gravidas.¹³ Stroke volume and cardiac output estimations correlated best, followed by pulmonary artery and ventricular filling pressures. This study provided some of the earliest evidence in support of noninvasive approaches to hemodynamic evaluation in pregnancy.¹ The practical application of noninvasive monitoring was demonstrated in a follow-up study by Belfort et al of 14 patients where all but 2 critically ill patients were able to be successfully managed without the use of a pulmonary artery catheter.¹⁴ Transesophageal Doppler can measure descending aortic flow used to estimate trends stroke volume and cardiac output which can be helpful during a code, particularly in the operating room.¹⁵

Stroke volume estimation by 2D echocardiogram techniques have been validated in pregnancy¹⁶ and this value is then multiplied by the heart rate to obtain the cardiac output (remember CO = SV × HR). In the future, 3D echocardiogram techniques may replace 2D as they provide more accurate geometric assessment of the cardiac chambers and reduce inherent errors based on geometric assumptions made using 2D echocardiography.^17-19 Also, the accuracy between measurements of chamber volume and correlation to pressure depend on the compliance of the chamber measured. For example, if diastolic dysfunction is present (or conditions where the chamber does not relax well: ie, tachycardia, ischemia, sepsis, high intrathoracic pressure), the end-diastolic volume (EDV) measured by echocardiogram would be reduced but filling pressures measured by PAC would be elevated. The PAC measurement alone might cause one to think this patient has adequate volume but these pressures are elevated in the context of poor wall compliance. Longitudinal studies have been performed to assess baseline 2D echocardiographic variables in pregnancy and postpartum (Table 1-4).

Arterial pressure waveform (APWF) monitoring uses the principle that stroke volume can be estimated from the morphology of the arterial pressure curve and has been validated for use in pregnant patients. Accuracy of APWF monitoring is limited by arrhythmias, use of intra-aortic balloon pump (IABP), septic shock, and liver failure.¹

Thoracic electrical bioimpedance technology has also been validated for use in pregnancy; however, estimations may be less reliable in obese patients and results are affected by maternal position.^21,22 One advantage of bioimpendance methods is the ability to follow beat-to-beat variations in CO.²³

Expected physiologic changes of pregnancy have been studied since the 1950s. Studies consistently demonstrate increased preload and volume over the course of gestation, starting as early as 7 to 8 weeks of pregnancy. Other changes include increased cardiac output, with stroke volume mainly driving this increase. The volume loading or “gestational hypervolemia” that occurs in pregnancy increases stroke volume and happens mainly over the first two trimesters of a normal pregnancy. Toward the end of the third trimester, cardiac output reaches a plateau and remains relatively constant until the time of labor.²³ There are dramatic hemodynamic changes that normally occur during labor which will be reviewed later in this chapter (Table 1-5).

There tends to be little change overall in PAOP and CVP in pregnancy despite significant increases in volume.¹ This is because of progesterone mediated decreases in systemic vascular resistance (SVR) and pulmonary vascular resistance (PVR). With less resistance, the circulatory system can accommodate more volume without a real appreciable change in pressure.

Cardiac remodeling occurs over time in pregnancy. This is evidenced by EKG changes that reflect cellular hypertrophy, particularly of the left ventricle and slight change in cardiac axis toward left deviation,^1,23 which is partially an effect of the encroaching gravid uterus on the diaphragm. All chambers become enlarged, in particular the left atrium (LA).²⁰ This enlargement of the LA and stretching of myocytes can trigger supraventricular arrhythmias, particularly in predisposed patients (thyroid disease, congenital heart disease with or without repair). Annular diameters of the valves increase.²⁰ A flow murmur is normal in pregnancy. In over 90% of women, mild pulmonary and tricuspid regurgitation occurs which is not considered pathologic.^24,25 Up to 33% of women will show evidence of clinically insignificant mitral regurgitation as well.²⁴

Effect of Position Change on Maternal Hemodynamics

A 20-week sized uterus (at the level of the umbilicus) is large enough to compress the inferior vena cava (IVC) which reduces venous return leading to diminished stroke volume and cardiac output; ultimately resulting in decreased blood pressure. This phenomenon seen in pregnancy is called supine hypotension or supine hypotensive syndrome. Positional effects on cardiac output become more impressive as the pregnancy progresses. There can be up to a 30% decrease in cardiac output in the supine position as compared to the lateral recumbent position. A lateral tilt (or uterine displacement) of at least 15% is sufficient to relieve pressure from the gravid uterus on the IVC and restore normal cardiac output. This should be done anytime augmentation of cardiac output is necessary. During cardiac resuscitation, however, a 30% tilt is recommended for maximum augmentation of cardiac output.²⁶