2. Because of the increased risk of postpartum hemorrhage in women with heart disease who are anticoagulated, the introduction or reintroduction of warfarin should be delayed until at least 2 days postpartum. Meticulous monitoring of anticoagulation is essential.
3. It is essential that any drugs used to treat a woman’s cardiac condition are checked for safety in the newborn infant if she is breastfeeding. This is particularly important if the infant was born prematurely.
4. Prior to discharge from hospital, it is essential to discuss plans for future contraception and arrange timely follow up with the cardiology team.
The puerperium is a dangerous time for women with heart disease, especially those with cardiomyopathy or pulmonary hypertension (PH). This is the time of highest mortality and morbidity risk. In a short space of time, many changes affect cardiac function. Some of these hemodynamic changes are predictable (e.g. “autotransfusion” of blood into the systemic circulation as the uterus contracts), but others are unpredictable (e.g. a variable amount of blood loss per vaginam at the time of, or after, delivery). Although labor and much of its cardiovascular challenges can be avoided by delivering a woman operatively, the events of the puerperium are faced by everyone.
There have been several studies of hemodynamic changes occurring during pregnancy and the puerperium, using a variety of different invasive and noninvasive methodologies.[1–7] However, not many of these studies have concentrated on events happening in the first 48 h after delivery. Only one study followed women from preconception through pregnancy to 4 months postpartum ; others have used a variety of time points after pregnancy as surrogates for the nonpregnant state.
Pregnancy is characterized by a volume-loaded state, with increased cardiac output due to increased stroke volume and heart rate (HR) and increased preload, as demonstrated by increased ventricular end-diastolic dimensions.[1,3,7] Afterload is reduced, with a reduction in both systemic and pulmonary vascular resistances. Intrinsic myocardial contractility is increased in pregnancy.[3,6] Left ventricular wall thickness and mass increases in pregnancy,[6–8] with the left ventricle becoming more spherical in shape in the third trimester. Filling pressures on the left side of the heart (pulmonary capillary wedge pressure [PCWP]) and on the right side of the heart (central venous pressure) are unchanged, as is pulmonary artery pressure. Colloid osmotic pressure (COP) and the COP–PCWP gradient are low in late pregnancy, contributing to the vulnerability of women to pulmonary edema around the time of delivery.[4,9]
During active labor, cardiac output increases by 15–30% during contractions; this is associated with a 25–30% increase in stroke volume, an increase in blood pressure, and a decrease in HR.[10,11] In the second stage of labor, cardiac output increases, reaching levels 40% higher than those recorded before labor, except in women with adequate regional anesthesia for whom cardiac output is hardly changed. Active pushing, involving a repeated Valsalva maneuver, results in a reduction in cardiac output, which overshoots as the HR rises. Pain and anxiety increase cardiac output.
Changing from the left lateral to the supine or upright posture diminishes cardiac output by 9% and 18%, respectively. Impedance cardiography studies of the effects of position change in late pregnancy found subtle changes in stroke volume, but no change in the parameters of cardiac contractility. This was attributed to increased venous return in the left lateral position and to the heart functioning at the peak of the Frank Starling curve.
Cardiac output significantly increases immediately after delivery, typically 60% above prelabor values, regardless of the type of analgesia used. Not all of the rise in cardiac output can be explained by tachycardia. In one study, 10 min after delivery of the baby, stroke volume was found to be increased by 50% over prelabor values.
In a study by Kjeldsen, at 1 h and 2 h postdelivery, cardiac output was on average 22% higher, stroke volume was 38% higher, but HR was lower than before the onset of labor. Serial central venous pressure measurements in this study showed significantly increased values after delivery. This was attributed to “autotransfusion” of blood from the uteroplacental bed following delivery.
High cardiac output is preserved for at least 48 h and, despite a reduction in HR at this time, stroke volume increases.[1,14] There is an increase in left ventricular end-diastolic dimension, suggesting increased venous return due to increased circulating volume. A recent echocardiographic study found evidence of left ventricular systolic and diastolic dysfunction in healthy women 48 h postpartum, and also higher levels of the cardiac biomarkers adrenomedullin, soluble fms-like tyrosine kinase-1 (FLT-1; also known as vascular endothelial growth factor receptor 1 [VEGFR-1]) and high sensitivity C-reactive protein compared with values obtained in healthy nonpregnant women.
Cardiovascular parameters measured 2 weeks after delivery show a reduction in cardiac output, stroke volume, and HR. Left ventricular end-diastolic dimension, ejection fraction, and circumferential shortening are also reduced compared with values recorded in late pregnancy or the early puerperium. These changes suggest not only that the circulating volume declines but also that there is a reduction in myocardial contractility.
By 6 weeks, some further reduction in cardiac output, stroke volume, and HR has occurred, along with increased systemic and pulmonary vascular resistance. Myocardial contractility is reduced further. Left ventricular wall thickness decreases. It is unclear when or even whether cardiovascular parameters reach a true “nonpregnant” state (i.e. the same as in women who have never been pregnant). Robson et al. followed a cohort of 15 women for 6 months from pregnancy, performing Doppler and M-mode echocardiography studies at 38 weeks of gestation and then at 2, 6, 12, and 24 weeks after delivery. Even at 24 weeks postpartum, many cardiovascular parameters were significantly different from those found in an age-matched control group of women, none of whom were using hormonal contraception. In particular, postnatal women had higher left ventricular wall thickness, dilatation of aortic and pulmonary valve areas, higher stroke volume, and reduced indices of left ventricular contractility. In the only hemodynamic study to follow the same cohort of women through from preconception to 4 months postpartum, there were significantly lower brachial and central systolic blood pressures postpartum; however, HR, cardiac output, stroke volume, and peripheral vascular resistance had returned to prepregnancy baseline values.
The concept of “autotransfusion” of blood from the uteroplacental bed into the systemic circulation once the uterus contracts tonically after delivery of the placenta is an important one. Studies have estimated the magnitude of these blood volume changes to be 300–500 ml. Similar volumes of blood are squeezed from the uteroplacental bed during labor contractions. There is a rise in central venous pressure measured during contractions and in the first 2 h following delivery, consistent with these acute changes in circulating blood volume. Postdelivery “autotransfusion” is a significant hemodynamic phenomenon, despite blood losses per vaginam during delivery.
There is also increased venous return after delivery due to relief of aortocaval compression by the uterus.
Total blood volume may be diminished by vaginal blood loss at the time of delivery or during the following hours and days. It is not uncommon for brisk hemorrhage to arise from uterine atony, or for substantial losses to occur from vaginal and perineal trauma sustained during delivery.
The increase in plasma volume noted during pregnancy diminishes after delivery. Most of the changes are likely to occur during the first 2 weeks, in parallel with the diminution of cardiac output. Diuresis is the response to decreased activity of the renin–angiotensin–aldosterone axis following pregnancy. Extravascular fluid is also mobilized, enters the circulation, and is then lost to the body in urine.
Many of the drugs used at the time of delivery have cardiovascular effects: Ergometrine causes vasoconstriction and transiently elevated blood pressure, oxytocin (Syntocinon®; Alliance, Chippenham, UK) given as an intravenous bolus may cause hypotension, while oxytocin given as an ongoing intravenous infusion has antidiuretic hormone effects, leading to water retention and hyponatremia. In the early puerperium, there may also be residual effects from anesthetic agents, particularly the sympathetic blockade created by regional anesthesia. Intravenous fluids may also have been given during labor, as vehicles for drug administration, for hydration purposes, or in association with regional anesthesia.
Women are at their most vulnerable time for thromboembolism peridelivery. There is maximal platelet activation and fibrin formation at the time of delivery. Additionally, by 3 h postdelivery, there is continued evidence of clotting activity and also maximum fibrinolysis. Physical factors such as immobility and impaired venous drainage contribute to the risk of venous thromboembolism; other factors (e.g. pyrexia and dehydration) may also be important.
Risks during the puerperium for women with heart disease
Healthy women tolerate the dramatic hemodynamic changes of the puerperium, mostly without problems. However, women with underlying heart disease may be at risk of one or more of the following adverse events:
alteration in cardiac shunting of blood
The particular risks depend on the underlying cardiac condition, but can also be influenced by intercurrent disease, particularly preeclampsia, thrombophilia, diabetes, and sepsis.
Fluid flow into the pulmonary interstitium depends on capillary permeability and the difference between hydrostatic and osmotic pressures within the capillaries. As discussed above, the COP–PCWP gradient is low in late pregnancy ; gradients of ≤4 mmHg are associated with pulmonary edema. Factors that increase pulmonary capillary wedge pressure, reduce COP, or increase capillary permeability predispose to the development of pulmonary edema. These may be inherent to cardiac disease (e.g. mitral stenosis causing increased pulmonary capillary wedge pressure) or consequent on other conditions (e.g. preeclampsia can be associated with increased pulmonary capillary wedge pressure, reduced COP, and increased capillary permeability).
Fluid shifts occurring soon after delivery as the uteroplacental circulation constricts or later as accumulated extracellular fluid is mobilized into the circulation can give rise to systemic hypertension. High circulating catecholamine levels, in response to anxiety or pain, result in elevated blood pressure and cardiac output. Intercurrent preeclampsia is associated with labile and sometimes extremely elevated blood pressure, due to increased vascular reactivity to circulating pressor agents. Women with preeclampsia often have reduced plasma volumes compared with healthy women in late gestation; thus, they often have much higher systemic vascular resistance (SVR) than would be typical in pregnancy.
Systemic vasoconstriction or vasodilatation in response to blood loss or the sympathetic blockade of regional anesthesia alters the direction of blood flow through intracardiac shunts. This affects venous return, cardiac output, and pulmonary blood flow. A sudden fall in ventricular output may cause cardiogenic shock.[19,20] There may be subtle changes in the balance of systemic and pulmonary vascular resistances following delivery due to the loss of aortocaval compression, improvement of venous return to the heart, and acute increase in plasma volume; these may be enough to affect direction of shunt flow in women with PH and a large intracardiac shunt.
This can arise because of alteration in the direction of flow across an intracardiac shunt, such that blood flows right to left in the heart and does not enter the pulmonary circulation. Cyanosis can also be due to worsening PH because pulmonary vascular resistance increases gradually in the weeks after delivery or acutely in response to hypoxia, hypercarbia, acidosis, stress, or pain. Thrombosis in the small vessels of the pulmonary arterial system or multiple pulmonary emboli from a systemic venous thrombosis will also increase pulmonary vascular resistance.
When a deep vein thrombosis has formed, pulmonary embolization may occur as an acute event or multiple, recurrent events. In someone with an intracardiac shunt, there is the possibility of systemic embolization, causing cerebral transient ischemic attacks or infarction.[22,23] Clots may also develop in dilated or poorly contracting heart chambers, further compromising cardiac function. Women with cyanotic heart disease are commonly erythrocytotic, may have increased blood viscosity, and are at risk of thromboembolism.
Effective ventricular filling and emptying depends on cardiac rate. Blood flow into the coronary arteries happens during diastole and can be impaired if there is a persistent tachycardia, especially when there is ventricular hypertrophy. Acute rhythm disturbances are more likely when the cardiac chambers are dilated, but may also be the consequence of reentry phenomena, ischemic heart disease, or the long QT syndrome. Any acute or persistent tachyarrhythmia in the puerperium can impair cardiac output, cause syncope, angina, or precipitate cardiac arrest.
Cesarean delivery is associated with a higher incidence (approximately 3%) of bacteremia compared with vaginal delivery (0.1%). The highest risks of bacteremia are associated with prolonged rupture of the membranes, chorioamnionitis, pyelonephritis, or endometritis, the latter particularly following cesarean delivery. Thus, most severe sepsis (approximately 80%) in obstetric patients happens in the postpartum period. The risks for women with cardiac disease are those of the general hemodynamic disturbance associated with a pyrexial illness (e.g. vasodilatation, tachycardia), increased risk of pelvic vein thromboembolism, and risks of bacterial endocarditis in those with susceptible conditions (e.g. valvular disease, prosthetic valves, ventricular septal defects).
Myocardial infarction, although uncommon, is most likely to happen during labor or in the postpartum period.[24,25] The mechanism more often involves coronary artery dissection, spasm, or thrombus than atheromatous disease. Hemodynamic changes around the time of delivery exert maximal sheer stresses in the aorta and coronary vessels, which may also have weakened collagen at this time. The risks of dissection are greatest in women with a known connective tissue disorder, e.g. Marfan or Ehlers–Danlos syndrome. Acute myocardial infarction is more common in older mothers with hypertensive comorbidities, either pre-existing or pregnancy induced. Exposure to ergometrine or abuse of cocaine has also been linked to coronary artery spasm and dissection. Ischemia may also complicate the puerperium in women with ventricular outflow obstruction and ventricular wall hypertrophy, cardiac transplantation, coronary arteritis, or collagen vascular diseases, and those with known occlusive coronary artery disease.[19,27]
Deaths from Eisenmenger syndrome, which has a 30–50% maternal mortality rate, usually occur within the first week after delivery. Women with PH, however, are as likely to die antenatally or in the late puerperium as in the first weeks following delivery. Cardiomyopathy-related deaths tend to happen many months after delivery and some occur much later.[29,30] Deaths from myocardial infarction happen both early and late in the puerperium.[19,26]