Hacker & Moore's Essentials of Obstetrics and Gynecology: With STUDENT CONSULT Online Access,5th ed.

Chapter 6

Maternal Physiologic and Immunologic Adaptation to Pregnancy

Brian J. Koos, Daniel A. Kahn, Ozlem Equils

Maternal physiologic adjustments to pregnancy are designed to support the requirements of fetal homeostasis and growth without unduly jeopardizing maternal well-being. This is accomplished by remodeling maternal systems to deliver energy and growth substrates to the fetus and to remove inappropriate heat and waste products. There appears to be a privileged immunologic sanctuary for the fetus and placenta during pregnancy.

image Normal Values in Pregnancy

The normal values for several hematologic, biochemical, and physiologic indices during pregnancy differ markedly from those in the nonpregnant range and may also vary according to the duration of pregnancy. These alterations are shown in Table 6-1.



image Cardiovascular System


The hemodynamic changes associated with pregnancy are summarized in Table 6-2. Retention of sodium and water during pregnancy accounts for a total body water increase of 6 to 8 L, two thirds of which is located in the extravascular space. The total sodium accumulation averages 500 to 900 mEq by the time of delivery. The total blood volume increases by about 40% above nonpregnant levels, with wide individual variations. The plasma volume rises as early as the 6th week of pregnancy and reaches a plateau by about 32 to 34 weeks’ gestation, after which little further change occurs. The increase averages 50% in singleton pregnancies and approaches 70% with a twin gestation. The red blood cell mass begins to increase at the start of the second trimester and continues to rise throughout pregnancy. By the time of delivery, it is 20% to 35% above nonpregnant levels. The disproportionate increase in plasma volume compared with the red cell volume results in hemodilution with a decreased hematocrit reading, sometimes referred to as physiologic anemia of pregnancy. If iron stores are adequate, the hematocrit tends to rise from the second to the third trimester.



Amount of Change


Arterial blood pressures



↓ 4-6 mm Hg

All bottom at 20-24 wk, then rise gradually to prepregnancy values at term


↓ 8-15 mm Hg


↓ 6-10 mm Hg


Heart rate

↑ 12-18 beats/min

1st, 2nd, 3rd trimesters

Stroke volume

↑ 10%-30%

1st and 2nd trimesters, then stable until term

Cardiac output

↑ 33%-45%

Peaks in early 2nd trimester, then stable until term

Data from Main DM, Main EK: Obstetrics and Gynecology: A Pocket Reference. Chicago, Year Book, 1984, p 18.

Cardiac output rises by the 10th week of gestation; it reaches about 40% above nonpregnant levels by 20 to 24 weeks, after which there is little change. The rise in cardiac output, which peaks while blood volume is still rising, reflects increases mainly in stroke volume and, to a lesser extent, in heart rate. With twin and triplet pregnancies, the changes in cardiac output are greater than those seen with singleton pregnancies.

The cardiovascular responses to exercise are altered during pregnancy. For any given level of exercise, oxygen consumption is higher in pregnant than in nonpregnant women. Similarly, the cardiac output for any level of exercise is also increased during pregnancy compared with that seen in a nonpregnant state, and the maximum cardiac output is reached at lower levels of exercise. It is not clear that any of the changes in hemodynamic responses to exercise are detrimental to mother and fetus, but it suggests that maternal cardiac reserves are lowered during pregnancy and that shunting of blood away from the uterus might occur during or after exercise.


Systolic pressure falls only slightly during pregnancy, whereas diastolic pressure decreases more markedly; this reduction begins in the first trimester, reaches its nadir in mid-pregnancy, and returns toward nonpregnant levels by term. These changes reflect the elevated cardiac output and reduced peripheral resistance that characterize pregnancy. Toward the end of pregnancy, vasoconstrictor tone, and with it blood pressure, normally increases. The normal, modest rise of arterial pressure as term approaches should be distinguished from the development of pregnancy-induced hypertension or preeclampsia. Pregnancy does not alter central venous pressures.

Blood pressure, as measured with a sphygmomanometer cuff around the brachial artery, varies with posture. In late pregnancy, arterial pressure is higher when the gravid woman is sitting compared with lying supine. When elevations in blood pressure are clinically detected during pregnancy, it is customary to repeat the measurement with the patient lying on her side. This practice usually introduces a systematic error. In the lateral position, the blood pressure cuff around the brachial artery is raised about 10 cm above the heart. This leads to a hydrostatic fall in measured pressure, yielding a reading about 7 mm Hg lower than if the cuff were at heart level, as occurs during sitting or supine measurements.


As pregnancy progresses, the enlarging uterus displaces and compresses various abdominal structures, including the iliac veins and inferior vena cava (and probably also the aorta), with marked effects. The supine position accentuates this venous compression, producing a fall in venous return and hence cardiac output. In most gravid women, a compensatory rise in peripheral resistance minimizes the fall in blood pressure. In up to 10% of gravid women, however, a significant fall occurs in blood pressure accompanied by symptoms of nausea, dizziness, and even syncope. This supine hypotensive syndrome is relieved by changing position to the side. The expected baroreflexive tachycardia, which normally occurs in response to other maneuvers that reduce cardiac output and blood pressure, does not accompany caval compression. In fact, bradycardia is often associated with the syndrome.

The venous compression by the gravid uterus elevates pressure in veins that drain the legs and pelvic organs, thereby exacerbating varicose veins in the legs and vulva and causing hemorrhoids.The rise in venous pressure is the major cause of the lower extremity edema that characterizes pregnancy. The hypoalbuminemia associated with pregnancy also shifts the balance of the other major factor in the Starling equation (colloid osmotic pressure) in favor of fluid transfer from the intravascular to the extracellular space. Because of venous compression, the rate of blood flow in the lower veins is also markedly reduced, causing a predisposition to thrombosis.The various effects of caval compression are somewhat mitigated by the development of a paravertebral collateral circulation that permits blood from the lower body to bypass the occluded inferior vena cava.

During late pregnancy, the uterus can also partially compress the aorta and its branches. This is thought to account for the observation in some patients of lower pressure in the femoral artery compared with that in the brachial artery. This aortic compression can be accentuated during uterine contractions and may be a cause of fetal distress when a patient is in the supine position. This phenomenon has been referred to as the Posiero effect. Clinically, it can be suspected when the femoral pulse is not palpable.


Blood flow to most regions of the body increases and reaches a plateau relatively early in pregnancy. Notable exceptions occur in the uterus, kidney, breasts, and skin, in each of which blood flow increases with gestational age. Two of the major increases (those to the kidney and to the skin) serve purposes of elimination: the kidney of waste material and the skin of heat. Both processes require plasma rather than whole blood, which gives point to the disproportionate increase of plasma over red blood cells in the blood expansion.

Early in pregnancy, renal blood flow increases to levels about 30% above nonpregnant levels and remains unchanged as pregnancy advances. This change accounts for the increased creatinine clearance and lower serum creatine level. Engorgement of the breasts begins early in gestation, with mammary blood flow increasing 2 to 3 times in later pregnancy. The skin blood flow increases slightly during the third trimester, reaching 12% of cardiac output.

There is little information on the distribution of blood flow to other organ systems during pregnancy. The uterine blood flow increases from about 100 mL/min in the nonpregnant state (2% of cardiac output) to about 1200 mL/min (17% of cardiac output) at term. Uterine blood flow and thus gas and nutrient transfer to the fetus are vulnerable. When maternal cardiac output falls, blood flow to the brain, kidneys, and heart is supported by a redistribution of cardiac output, which shunts blood away from the uteroplacental circulation. Similarly, changes in perfusion pressure can lead to decreases in uterine blood flow. Because the uterine vessels are maximally dilated during pregnancy, little autoregulation can occur to improve uterine blood flow.


The precise mechanisms accounting for the cardiovascular changes in pregnancy have not been fully elucidated. The rise in cardiac output and fall in peripheral resistance during pregnancy may be explained in terms of the circulatory response to an arteriovenous shunt, represented by the uteroplacental circulation. The elevations in cardiac output and uterine blood flow follow different time courses in pregnancy, however, with the former reaching its maximum in the second trimester and the latter increasing to term.

A unifying hypothesis suggests that the elevations in circulating steroid hormones, in combination with increases in production of aldosterone and vasodilators such as prostaglandins, atrial natriuretic peptide, nitric oxide, and probably others, reduce arterial tone and increase venous capacitance. These changes, along with the development of arteriovenous shunts, appear responsible for the increase in blood volume and the hyperdynamic (high-flow, low-resistance) circulation of pregnancy. The same hormonal changes cause relaxation in the cytoskeleton of the maternal heart, which allows the end-diastolic volume (and stroke volume) to increase.


Plasma volume expands proportionately more than red blood cell volume, leading to a fall in hematocrit. Optimal pregnancy outcomes are generally achieved with a maternal hematocrit of 33% to 35%. Hematocrit readings below about 27% or above about 39% are associated with less favorable outcomes. Despite the relatively low “optimal” hematocrit, the arteriovenous oxygen difference in pregnancy is below nonpregnant levels. This supports the concept that the hemoglobin concentration in pregnancy is more than sufficient to meet oxygen-carrying requirements.

Pregnancy requires about 1 g of elemental iron: 0.7 g for mother and 0.3 g for the placenta and fetus. A high proportion of women in the reproductive age group enter pregnancy without sufficient stores of iron to meet the increased needs of pregnancy.

image Respiratory System

The major respiratory changes in pregnancy involve three factors: the mechanical effects of the enlarging uterus, the increased total body oxygen consumption, and the respiratory stimulant effects of progesterone.


The changes in lung volume and capacities associated with pregnancy are detailed in Table 6-3. Assessment of mechanical changes during pregnancy reveals that the diaphragm at rest rises to a level of 4 cm above its usual resting position. The chest enlarges in transverse diameter by about 2.1 cm. Simultaneously, the subcostal angle increases from an average of 68.5 degrees to 103.5 degrees during the latter part of gestation. The increase in uterine size cannot completely explain the changes in chest configuration because these mechanical changes occur early in gestation.




Change in Pregnancy

Respiratory rate


No significant change

Tidal volume

The volume of air inspired and expired at each breath

Progressive rise throughout pregnancy of 0.1-0.2 L

Expiratory reserve volume

The maximum volume of air that can be additionally expired after a normal expiration

Lowered by about 15% (0.55 L in late pregnancy compared with 0.65 L postpartum)

Residual volume

The volume of air remaining in the lungs after a maximum expiration

Falls considerably (0.77 L in late pregnancy compared with 0.96 L postpartum)

Vital capacity

The maximum volume of air that can be forcibly inspired after a maximum expiration

Unchanged, except for possibly a small terminal diminution

Inspiratory capacity

The maximum volume of air that can be inspired from resting expiratory level

Increased by about 5%

Functional residual capacity

The volume of air in lungs at resting expiratory level

Lowered by about 18%

Minute ventilation

The volume of air inspired or expired in 1 min

Increased by about 40% as a result of the increased tidal volume and unchanged respiratory rate

Data from Main DM, Main EK: Obstetrics and Gynecology: A Pocket Reference. Chicago, Year Book, 1984, p 14.

As pregnancy progresses, the enlarging uterus elevates the resting position of the diaphragm. This results in less negative intrathoracic pressure and a decreased resting lung volume; that is, a decrease in functional residual capacity (FRC). The enlarging uterus produces no impairment in diaphragmatic or thoracic muscle motion. Hence, the vital capacity (VC) remains unchanged. These characteristics—reduced FRC with unimpaired VC—are analogous to those seen in a pneumoperitoneum and contrast with those seen in severe obesity or abdominal binding, where the elevation of the diaphragm is accompanied by decreased excursion of the respiratory muscles. Reductions in both the expiratory reserve volume and the residual volume contribute to the reduced FRC.


Total body oxygen consumption increases about 15% to 20% in pregnancy. About half of this increase is accounted for by the uterus and its contents. The remainder is accounted for mainly by increased maternal renal and cardiac work. Smaller increments are due to greater breast tissue mass and to increased work of the respiratory muscles.

In general, a rise in oxygen consumption is accompanied by cardiorespiratory responses that facilitate oxygen delivery (i.e., by increases in cardiac output and alveolar ventilation). To the extent that elevations in cardiac output and alveolar ventilation keep pace with the rise in oxygen consumption, the arteriovenous oxygen difference and the arterial partial pressure of carbon dioxide (PCO2), respectively, remain unchanged. In pregnancy, the elevations in both cardiac output and alveolar ventilation are greater than those required to meet the increased oxygen consumption. Hence, despite the rise in total body oxygen consumption, the arteriovenous oxygen difference and arterial PCO2 both fall. The fall in PCO2 (to 27-32 mm Hg), by definition, indicates hyperventilation.

The rise in minute ventilation reflects an approximate 40% increase in tidal volume at term; the respiratory rate does not change during pregnancy. During exercise, pregnant subjects show a 38% increase in minute ventilation and a 15% increase in oxygen consumption above comparable levels for postpartum subjects.

When injected into normal nonpregnant subjects, progesterone increases ventilation. The central chemoreceptors become more sensitive to CO2 (i.e., the curve describing the ventilatory response to increasing CO2 has a steeper slope). Such increased respiratory sensitivity to CO2 is characteristic of pregnancy and probably accounts for the hyperventilation of pregnancy.

In summary, both at rest and with exercise, minute ventilation and, to a lesser extent, oxygen consumption are increased during pregnancy over the nonpregnant control values. The respiratory stimulating effect of progesterone is probably responsible for the disproportionate increase in minute ventilation over oxygen consumption.


The hyperventilation of pregnancy results in a respiratory alkalosis. Renal compensatory bicarbonate excretion leads to a final maternal blood pH between 7.40 and 7.45. During labor (without conduction anesthesia), the hyperventilation associated with each contraction produces a further transient fall in PCO2. By the end of the first stage of labor, when cervical dilation is complete, a decrease in arterial PCO2 persists, even between contractions.

In general, when alveolar PCO2 falls during hyperventilation, alveolar partial pressure of oxygen (PO2) shows a corresponding rise, leading to a rise in arterial PO2. In the first trimester, the mean arterial PO2 may be 106 to 108 mm Hg. There is a slight downward trend in arterial PO2 as pregnancy proceeds. This reflects, at least in part, an increased alveolar-arterial gradient, possibly resulting from the decrease in FRC discussed previously, which leads to a ventilation-perfusion mismatch.


In general, airway resistance is unchanged or even decreased in pregnancy. Despite the absence of obstructive or restrictive effects, dyspnea is a common symptom in pregnancy. Some studies have suggested that dyspnea may be experienced at some time during pregnancy by as many as 60% to 70% of women. Although the mechanism has not been established, the dyspnea of pregnancy may involve the increased sensitivity and lowered threshold to PCO2.

image Renal Physiology


The urinary collecting system, including the calyces, renal pelves, and ureters, undergoes marked dilation in pregnancy, as is readily seen on intravenous urograms. It begins in the first trimester, is present in 90% of women at term, and may persist until the 12th to 16th postpartum week. Progesterone appears to produce smooth muscle relaxation in various organs, including the ureter. As the uterus enlarges, partial obstruction of the ureter occurs at the pelvic brim in both the supine and the upright positions. Because of the relatively greater effect on the right side, some have ascribed a role to the dilated ovarian venous plexus. Ovarian venous drainage is asymmetric, with the right vein emptying into the inferior vena cava and the left into the left renal vein.


Renal plasma flow and the glomerular filtration rate (GFR) increase early in pregnancy, with maximum plateau elevations of at least 40% to 50% above nonpregnant levels by mid-gestation, and then remain unchanged to term. As was true for cardiac output, renal blood flow and GFR (clinically measured as the creatinine clearance) reach their peak relatively early in pregnancy, before the greatest expansion in intravascular and extracellular volume occurs. The elevated GFR is reflected in lower serum levels of creatinine and urea nitrogen, as noted in Table 6-1.

Pregnancy is associated with large reductions in resistance in the afferent and efferent arterioles of the renal arteries, which appears to involve vasorelaxation induced by relaxin, endothelin, and nitric oxide. The resulting rise in renal plasma flow accounts for the hyperfiltration.


Although 500 to 900 mEq of sodium is retained during pregnancy, sodium balance is maintained with exquisite precision. Despite the large amounts of sodium consumed daily (100 to 300 mEq), only 20 to 30 mEq of sodium is retained every week. Pregnant women given high or low sodium diets are able to demonstrate decreases or increases in sodium tubular reabsorption, respectively, which maintain sodium and fluid balance.

Pregnant women also maintain fluid balance with no change in the concentrating or diluting ability of the kidney. Plasma osmolarity is reduced by about 10 mOsm/kg of water. Potassium metabolism during pregnancy is unchanged, although about 350 mEq of potassium is retained during pregnancy for fetoplacental development and expansion of maternal red cell mass.

The hyperventilation (low PaCO2) of pregnancy results in respiratory alkalosis, which is compensated by renal excretion of bicarbonate. As a result, maternal renal buffering capacity is reduced.


The maternal extracellular volume, which consists of intravascular and interstitial components, increases throughout pregnancy, leading to a state of physiologic extracellular hypervolemia. The intravascular volume, which consists of plasma and red cell components, increases about 50% during pregnancy. Maternal interstitial volume shows its greatest increase in the last trimester.

The magnitude of the rise in maternal plasma volume correlates with the size of the fetus; it is particularly marked in cases of multiple gestation. Multiparous women with poor reproductive histories show smaller increments in plasma volume and GFR when compared with those with a history of normal pregnancies and normal-sized babies.


Plasma concentrations of renin, renin substrate, and angiotensin I and II are increased during pregnancy. Renin levels remain elevated throughout pregnancy, with at least a portion of the renin circulating in a high-molecular-weight form.

The uterus, like the kidney, can produce renin, and extremely high concentrations of renin occur in the amniotic fluid. The physiologic role of uterine renin has not been established.

image Homeostasis of Maternal Energy Substrates

The metabolic regulation of energy substrates, including glucose, amino acids, fatty acids, and ketone bodies, is complex and interrelated.


In pregnancy, the insulin response to glucose stimulation is augmented. By the 10th week of normal pregnancy and continuing to term, fasting concentrations of insulin are elevated and those of glucose reduced. Until mid-gestation, these changes are accompanied by enhanced intravenous glucose tolerance (although oral glucose tolerance remains unchanged). Glycogen synthesis and storage by the liver increases, and gluconeogenesis is inhibited. Thus, during the first half of pregnancy, the anabolic actions of insulin are potentiated.

After early pregnancy, insulin resistance emerges, so glucose tolerance is impaired. The fall in serum glucose for a given dose of insulin is reduced compared with the response in earlier pregnancy. Elevation of circulating glucose is prolonged after meals, although fasting glucose remains reduced, as in early pregnancy.

A variety of humoral factors derived from the placenta have been suggested to account for the antiinsulin environment of the latter part of pregnancy. Perhaps the most important are cytokines and human placental lactogen (hPL), which antagonize the peripheral effects of insulin. An increase in levels of free cortisol and other hormones may also be involved in the insulin resistance of pregnancy.


The potentiated anabolic effects of insulin that characterize early pregnancy lead to the inhibition of lipolysis. During the second half of pregnancy, however, probably as a result of rising hPL levels, lipolysis is augmented, and fasting plasma concentrations of free fatty acids are elevated. Teleologically, the free fatty acids act as substrates for maternal energy metabolism, whereas glucose and amino acids cross the placenta to the fetus. In the humoral milieu of the second half of the pregnancy, the increased free fatty acids lead to ketone body (β-hydroxybutyrate and acetoacetate) formation. Pregnancy is thus associated with an increased risk for ketoacidosis, especially after prolonged fasting.

In the context of maternal lipid metabolism, the most dramatic lipid change in pregnancy is the rise in fasting triglyceride concentration.

image Placental Transfer of Nutrients

The transfer of substances across the placenta occurs by several mechanisms, including simple diffusion, facilitated diffusion, and active transport. Low molecular size and lipid solubility promote simple diffusion. Substances with molecular weights greater than 1000 Daltons, such as polypeptides and proteins, cross the placenta slowly, if at all.

Amino acids are actively transported across the placenta, making fetal levels higher than maternal levels. Glucose is transported by facilitated diffusion, leading to rapid equilibrium with only a small maternal-fetal gradient. Glucose is the main energy substrate of the fetus although amino acids and lactate may contribute up to 25% of fetal oxygen consumption. The degree and mechanism of placental transfer of these and other substances are summarized in Table 6-4.




Placental Transfer

Glucose homeostasis


Excellent—”facilitated diffusion”


Amino acids

Excellent—active transport


Free fatty acids (FFA)

Very limited—essential FFA only






No transfer



No transfer

Thyroid function

Thyroxine (T4)

Very poor—diffusion


Triiodothyronine (T3)



Thyrotropin-releasing hormone (TRH)



Thyroid-stimulating immunoglobulin (TSI)



Thyroid-stimulating hormone (TSH)

Negligible transfer




Adrenal hormones


Excellent transfer and active placental conversion of cortisol to cortisone beginning at mid-pregnancy



No transfer

Parathyroid function


Active transfer against gradient



Active transfer against gradient



Active transfer against gradient


Parathyroid hormone

Not transferred



Minimal passive transfer



Good—both passive and active transport from 7 wk gestation



No transfer

 At mid-gestation, placental 11-β hydroxysteroid dehydrogenase converts cortisol to cortisone.

Data from Main DM, Main EK: Obstetrics and Gynecology: A Pocket Reference. Chicago, Year Book, 1984, p 37.

image Other Endocrine Changes


The thyroid gland undergoes moderate enlargement during pregnancy. This is not due to elevation of thyroid-stimulating hormone (TSH), which remains unchanged. Placenta-derived human chorionic gonadotropin (hCG) has a TSH effect on the thyroid gland, which can result in abnormally low levels of TSH in the first trimester, when hCG concentrations are highest.

Circulating thyroid hormone exists in two primary active forms: thyroxine (T4) and triiodothyronine (T3). The former circulates in higher concentrations, is more highly protein bound, and is less metabolically potent than T3, for which it may serve as a prohormone. Circulating T4 is bound to carrier proteins, about 85% to thyroxine-binding globulin (TBG) and most of the remainder to another protein, thyroxine-binding prealbumin. It is believed that only the unbound fraction of the circulating hormone is biologically active. TBG is increased during pregnancy because the high estrogen levels induce increased hepatic synthesis. The body responds by raising total circulating levels of T4 and T3, and the net effect is that the free, biologically active concentration of each hormone is unchanged. Therefore, clinically, the free T4 index, which corrects the total circulating T4 for the amount of binding protein, is an appropriate measure of thyroid function, with the same normal range as in the nonpregnant state. Only minimal amounts of thyroid hormone cross the placenta.


Adrenocorticotropic hormone (ACTH) and plasma cortisol levels are both elevated from 3 months’ gestation to delivery. Although less so than thyroid hormones, circulating cortisol is also primarily bound to a specific plasma protein, corticosteroid-binding globulin (CBG), or transcortin. Unlike the level of thyroid hormones, the mean unbound level of cortisol is elevated in pregnancy; there is also some loss of the diurnal variation that characterizes its concentration in nonpregnant women.

image Weight Gain in Pregnancy

The average weight gain in pregnancy uncomplicated by generalized edema is 12.5 kg (28 lb). The components of this weight gain are indicated in Table 6-5. The products of conception constitute only about 40% of the total maternal weight gain.



image Placental Transfer of Oxygen and Carbon Dioxide


The placenta receives 60% of the combined ventricular output, whereas the postnatal lung receives a greater proportion of the cardiac output. Unlike the lung, which consumes little of the oxygen it transfers, a significant percentage of the oxygen derived from maternal blood at term is consumed by placental tissue. The degree of functional shunting of placental blood past exchange sites is about 10-fold greater than in the lung. A major cause of this functional shunting is probably a mismatch between maternal and fetal blood flow at the exchange sites, analogous to the ventilation-perfusion inequalities that occur in the lung.

The uteroplacental circulation subserves fetal gas exchange. Oxygen, carbon dioxide, and inert gases cross the placenta by simple diffusion. The rate of transfer is proportional to the difference in gas tension across the placenta and the surface area of the placenta; and the transfer rate is inversely proportional to diffusion distance between maternal and fetal blood. The placenta normally does not pose a significant barrier to respiratory gas exchange, unless it becomes separated (abruption placenta) or edematous (severe hydrops fetalis).

Figure 6-1 depicts the anatomic distribution of uterine and umbilical blood flow and O2 transfer across the placenta. A maternal shunt, which describes the fraction of blood shunted to the myoendometrium and is estimated to constitute 20% of uterine blood flow, is depicted. Similarly, a fetal shunt, which supplies blood to the placenta and fetal membranes and accounts for 19% of umbilical blood flow, is shown. The maternal-to-fetal PO2 and PCO2gradients are calculated from measurements of gas tensions in the uterine and umbilical arteries and veins. The umbilical vein of the fetus, like the pulmonary vein of the adult, carries the circulation’s most highly oxygenated blood. The umbilical venous PO2 of about 28 mm Hg is relatively low by adult standards. This relatively low fetal tension is essential for survival in utero because a high PO2 initiates physiologic adjustments (e.g., closure of the ductus arteriosus and vasodilation of the pulmonary vessels) that normally occur in the neonate but would be harmful in utero.


FIGURE 6-1 Placental transfer of oxygen and carbon dioxide. BE, base excess; Hb, hemoglobin.

(Adapted from Bonica JJ: Obstetric Analgesia and Anesthesia, 2nd ed. Amsterdam, World Federation of Societies of Anesthesiologists, 1980, p 29.)

Although not involved in respiratory gas exchange, fetal breathing movements are critically involved in lung development and in the development of respiratory regulation. Fetal breathing differs from that in the adult in that it is episodic, sensitive to fetal glucose concentrations, and inhibited by hypoxia. Because of its sensitivity to acute O2 deprivation, fetal breathing is used clinically as indicator of the adequacy of fetal oxygenation.


Most of the oxygen in blood is carried by hemoglobin in red blood cells. The maximum amount of oxygen carried per gram of hemoglobin, that is, the amount carried at 100% saturation, is fixed at 1.37 mL. The hemoglobin flow rates depend on blood flow rates and hemoglobin concentration. The uterine blood flow at term has been estimated at 700 to 1200 mL/min, with about 75% to 88% of this entering the intervillous space. The umbilical blood flow has been estimated at 350 to 500 mL/min, with more than 50% going to the placenta (see Figure 6-1).

The hemoglobin concentration of the blood determines its oxygen-carrying capacity, which is expressed in milliliters of oxygen per 100 mL of blood. In the fetus at or near term, the hemoglobin concentration is about 18 g/dL, and oxygen-carrying capacity is 20 to 22 mL/dL. Maternal oxygen-carrying capacity of blood, which is generally proportional to hemoglobin concentration, is lower than that of the fetus.

The affinity of hemoglobin for oxygen, which is reflected as the percent saturation at a given oxygen tension, depends on chemical conditions. As is illustrated in Figure 6-2when compared with that in nonpregnant adults, the binding of oxygen by hemoglobin is much greater in the fetus under standard conditions of PCO2, pH, and temperature. In contrast, maternal affinity is lower under these conditions, with 50% of hemoglobin saturated with O2 at a PO2 of 26.5 mm Hg (P50) for mother compared with 20 mm Hg for the fetus.


FIGURE 6-2 The oxygen dissociation curve for fetal blood compared with maternal blood. The central continuous curve (in red) is for normal adult blood under standard conditions. A vertical line at an oxygen partial pressure of 30 mm Hg divides the curves. The fetal curve normally operates below that level (to the left) and the maternal curve above it (to the right).

(Adapted from Hytten F, Chamberlain G [eds]: Clinical Physiology in Obstetrics, 2nd ed. Oxford, Blackwell, 1991, p 418.)

In vivo, the greater fetal temperature and lower pH shift the O2-dissociation curve to the right, while the lower maternal temperature and higher pH shift the maternal curve to the left. As a result, the O2-dissociation curves for the fetal and maternal blood are not too dissimilar at the site of placental transfer. Maternal venous blood probably has an O2 saturation of about 73% and a PO2 of about 36 mm Hg, and the corresponding values for blood in the umbilical vein are about 63% and 28 mm Hg. As the only source of O2 for the fetus, blood in the umbilical vein has a higher O2 saturation and PO2 than blood in the fetal circulation (Figure 6-3). In the presence of a low fetal arterial PO2, fetal oxygenation is maintained by a high rate of blood flow to fetal tissues, which is supported by a very high cardiac output. This feature, along with the lower P50 of fetal blood, results in normal O2 delivery to fetal organs.


FIGURE 6-3 The fetal circulation. Numbers represent approximate values of percent saturation of blood with oxygen in utero.

(Adapted from Parer JJ: Fetal circulation. In Sciarra JJ [ed]: Obstetrics and Gynecology, Vol 3: Maternal and Fetal Medicine. Hagerstown, MD, Harper & Row, 1984, p 2.)

The decrease in the affinity of hemoglobin for oxygen produced by a fall in pH is referred to as the Bohr effect. Because of the unique situation in the placenta, a double Bohr effect facilitates oxygen transfer from mother to fetus. When CO2 and fixed acids are transferred from fetus to mother, the associated rise in fetal pH increases the fetal red blood cell’s affinity for oxygen uptake. The concomitant reduction in maternal blood pH decreases oxygen affinity and promotes its unloading of oxygen from maternal red cells.

image Fetal Circulation

Several anatomic and physiologic factors must be noted in considering the fetal circulation (Table 6-6; see Figure 6-3).


Fetal Structure


Adult Remnant

Umbilical vein

Umbilicus/ductus venosus

Ligamentum teres hepatis

Ductus venosus

Umbilical vein/inferior vena cava (bypasses liver)

Ligamentum venosum

Foramen ovale

Right atrium/left atrium

Closed atrial wall

Ductus arteriosus

Pulmonary artery/descending aorta

Ligamentum arteriosum

Umbilical artery

Common iliac artery/umbilicus

Superior vesical arteries; lateral vesicoumbilical ligaments

Data from Main DM, Main EK: Obstetrics and Gynecology: A Pocket Reference. Chicago, Year Book, 1984, p 34.

The normal adult circulation is a series circuit with blood flowing through the right heart, the lungs, the left heart, the systemic circulation, and finally the right heart. In the fetus, the circulation is a parallel system with the cardiac outputs from the right and left ventricles directed primarily to different vascular beds. For example, the right ventricle, which contributes about 65% of the combined output, pumps blood primarily through the pulmonary artery, ductus arteriosus, and descending aorta. Only a small fraction of right ventricular output flows through the pulmonary circulation. The left ventricle supplies blood mainly to the tissues supplied by the aortic arch, such as the brain. The fetal circulation is a parallel circuit characterized by channels (ductus venosus, foramen ovale, and ductus arteriosus) and preferential streaming, which function to maximize the delivery of more highly oxygenated blood to the upper body and brain, less highly oxygenated blood to the lower body, and very low blood flow to the nonfunctional lungs.

The umbilical vein, carrying oxygenated (80% saturated) blood from the placenta to the fetal body, enters the portal system. A portion of this umbilical-portal blood passes through the hepatic microcirculation, where oxygen is extracted, and thence through the hepatic veins into the inferior vena cava. Most of the blood bypasses the liver through the ductus venosus, which directly enters the inferior vena cava, which also receives the unsaturated (25% saturated) venous return from the lower body. Blood reaching the heart through the inferior vena cava has an oxygen saturation of about 70%, which represents the most highly oxygenated blood in the heart. About one third of blood returning to the heart from the inferior vena cava preferentially streams across the right atrium, mixing with blood from the superior vena cava to the foramen ovale into the left atrium, where it mixes with the relatively meager pulmonary venous return. Blood flows from the left atrium into the left ventricle, and then to the ascending aorta.

The proximal aorta, carrying the most highly saturated blood leaving the heart (65%), gives off branches to supply the brain and upper body. Most of the blood returning through the inferior vena cava enters the right atrium, where it mixes with the unsaturated blood returning through the superior vena cava (25% saturated). Right ventricular outflow (O2 saturation of 55%) enters the aorta through the ductus arteriosus, and the descending aorta supplies the lower body with blood having less O2 saturation (about 60%) than that flowing to the brain and the upper body.

The role of the ductus arteriosus must be emphasized. Right ventricular output enters the pulmonary trunk, from which its major portion, owing to the high vascular resistance of the pulmonary circulation, bypasses the lungs by flowing through the ductus arteriosus to the descending aorta. Although the descending aorta supplies branches to the lower fetal body, the major portion of descending aortic flow goes to the umbilical arteries, which carry deoxygenated blood to the placenta.


The following changes occur after birth (see Table 6-6):

1. Elimination of the placental circulation, with interruption and eventual obliteration of the umbilical vessels

2. Closure of the ductus venosus

3. Closure of the foramen ovale

4. Gradual constriction and eventual obliteration of the ductus arteriosus

5. Dilation of the pulmonary vessels and establishment of the pulmonary circulation

The elimination of the umbilical circulation, closure of the vascular shunts, and establishment of the pulmonary circulation will change the vascular circuitry of the neonate from an “in parallel” system to an “in series” system.

image Immunology of Pregnancy

Nearly 60 years ago, Peter Medawar recognized the apparent paradox of the immunologic evasion of the semiallogenic fetus to maternal response. He proposed three hypotheses to explain this paradox: (1) anatomic separation of mother and fetus; (2) antigenic immaturity of the fetus; or (3) immunologic “inertness” (tolerance) of the mother. In the intervening years, it has become apparent that both the mother and her fetus are immunologically aware of one another and yet tolerance exists for the most part. Furthermore, while the maternal immune response during pregnancy is qualitatively different, pregnancy does not result in an overall maternal immunosuppression.

It is clear that the growth and development of a semiallogeneic conceptus within an immunologically competent mother depends on the manner in which pregnancy alters the immune regulatory mechanisms. Historically, attention in addressing the “Medawar paradox” has focused exclusively on the mother, but it is now known that mammalian fetuses are capable of mounting immune responses in utero. The interplay between the fetal and maternal immune systems is complex and is a current active area on investigation.


Mammalian (including human) immune systems have two fundamental responses: an early “innate” and a later more specific and robust adaptive response.

The innate immune system response is the first line of defense and includes surface barriers (mucosal immunity), saliva, tears, nasal secretions, perspiration, blood and tissue monocyte-macrophages, natural killer (NK) cells, endothelial cells, polymorphonuclear neutrophils, the complement system, dendritic cells, and the normal microbial flora. The adaptive immune system is composed of cell-mediated (T lymphocytes) and humoral (B lymphocytes-antibodies) responses. Activation of T and consequently B lymphocytes is critical for the development of lifelong memory immune responses.

Innate immune cells have evolutionary acquired mechanisms that recognize the foreign nature of the inciting antigen and mount a transient protection within hours. There is no need for major histocompatibility complex (MHC) molecules. The epithelial cell interaction with the antigens induces the release of cytokines and chemokines, which attract the macrophages, dendritic cells, and NK cells. Macrophages and neutrophils then engulf and lyse the pathogens and produce cytokines. NK cells play the key role in destroying the virally infected cells. Damaged epithelial cells lead to the activation of complements. Complements can directly kill the microbes by punching holes in their membrane and indirectly by opsonizing them, which facilitates their phagocytosis. Complements also promote the inflammatory cell recruitment. The cytokines released from the immune cells activate the vascular endothelial cells, increasing permeability, allowing immune effector cells to penetrate into the tissues.

The critical link between the innate immune response and the adaptive immune response is antigen presentation. Foreign proteins that are phagocytosed are processed intracellularly, and then expressed on the cell surface complexed with MHC II. Additionally, the presenting cells provide critical secondary signals (through cell surface molecules) that are permissive for appropriate T-cell activation. Among the most efficient antigen-presenting cells are dendritic cells.

Dendritic cells play a key role in alerting the adaptive immune responses. Immature dendritic cells engulf the pathogens, carry them to the lymph nodes, and present them to CD4+ T lymphocytes. Activated T cells develop surface receptors for specific foreign antigens and undergo clonal proliferation. Cytotoxic (activated) T cells can directly kill target cells expressing viral antigens together with MHC I. In contrast to antigens presented in the context of MHC II, a portion of all cellular proteins are expressed on the cell surface of all normal cells in the context of MHC I. By this mechanism, the immune system can determine whether a cell is producing self proteins or if the cell has been altered (e.g., by virus) to produce foreign proteins.

Once CD4+ T cells are activated, they can direct an immune response by secreting proteins (cytokines) that activate surrounding cells. By secreting interferon-γ and interleukin-2 (IL-2), a CD4+ T cell induces a cellular immune response through CD8+ “killer” T cells. By secreting IL-4 and IL-5, CD4+ T cells promote B cells to proliferate and differentiate for immunoglobulin (antibody) production. B cells exposed to antigen for the first time produce immunoglobulin M (IgM). As the affinity of the immunoglobulin (antibody) increases, the B cell undergoes a genetic rearrangement and may produce a variety of different antibodies. The most specific are usually of the IgG subtype. IgG crosses the placenta and will accumulate into the fetus.


The innate immune effector cells first arise from hematopoietic progenitors noted in the blood islands of the yolk sac. By 8 embryonic weeks, the fetal liver becomes the source of these cells, and by 20 weeks, the fetal bone marrow takes over.

Macrophage-like cells arise from the yolk sac around 4 weeks; by 16 weeks, a fetus has the same number of circulating macrophages as adults, but they are less functional. The fetus has fewer tissue macrophages. Immature granulocytes can be found in the fetal spleen and liver by 8 weeks. NK cells are detected in the liver by 8 to 13 weeks and complements 2 and 4 by 8 weeks. C1, 3, 5, 7, 9 are found in the serum by 18 weeks. Maternal complements do not cross placenta into the fetus. The complement system continues to mature after parturition, and adult levels are reached by 1 year of age. Skin, one of the main innate barriers, completes its development 2 to 3 weeks after birth.

The cellular component of the adaptive immunity, T cells, are also derived from hematopoietic progenitors that are first seen in the blood islands of the yolk sac by 8 weeks. To differentiate into activated T cells, they must first migrate to the thymus gland, a relatively large organ in the fetus, the sole function of which appears to be to nurture and develop T cells. After maturation, T cells develop into either CD4 or CD8 types according to the surface receptor expressed. By 16 weeks, the thymus contains T cells in proportion to those found in the adult. In the newborn, the proportion of CD4 helper T cells and CD8 T cells is similar to that in the adult. However interferon-γ production is less efficient in fetal CD4 helper T cells.

Fetal B cells are first detected in the liver by 8 weeks, and around the second trimester, B-cell production is mostly from the bone marrow. Fetal B cells secrete IgG or IgA during the second trimester, but IgM antibodies are not secreted until the third trimester. Cord IgM levels greater than 20 mg/dL suggest an intrauterine infection. Maternal IgG crosses the placenta as early as the late first trimester, but the efficiency of the transport is poor until 30 weeks. Significant passive immunity can be transferred to the fetus in this manner, and for this reason, premature infants are not as well protected by maternal antibodiesIgM, because of its larger molecular size, is unable to cross the placenta. The other immunoglobulins (IgA, IgD, and IgE) are also confined to the maternal compartment, but the fetus can make its own IgA and IgM.

Physiologically, newborns have higher neutrophil and lymphocyte counts. The neutrophil counts decrease by 1 week of age, whereas lymphocyte counts continue to rise. The proportion of lymphocytes and absolute lymphocyte counts are higher in neonates compared with adults.


Pregnancy poses a special immunologic problem. The embryo must implant and cause a portion (placenta) to invade the uterine lining in order to gain access to the maternal circulation for nutrition and gas exchange. The maintenance of the antigenically dissimilar fetus in the uterus of the mother is of primary importance in obstetrics. The total picture of immune regulation at the maternal-fetal interface is yet to be elucidated, but the following is a synopsis of the current level of understanding.

The primary sites of modulation of the maternal response are the uterus, regional lymphatics, and placenta. In the uterus, NK-cell–mediated inflammation is necessary for the appropriate attachment and penetration of the fertilized egg into the uterine wall and for early placental development, whereas increased suppressive T cells, the presence of molecules that inactivate the previously activated maternal lymphocytes (CTLA4), and the absence of B cells provide the needed immune quiescence to allow for successful pregnancy. The placenta and the membranes provide the key barrier in protecting the growing fetus from microbial pathogens and toxins circulating in the mother’s blood. Syncytiotrophoblast, which makes up the cell barrier between the fetal and maternal blood in the placenta, does not express classic self and nonself MHC I and II molecules. Deeper trophoblastic cells do not express MHC II, but some express MHC I and are not stimulatory. This allows protection from invading microbes but at the same time prevents the destruction of the fetus.

HLA-G suppresses the adaptive and innate immune responses in the placenta and promotes the release of antiinflammatory cytokines such as IL-10. The soluble forms of HLA-G are found in the blood of pregnant women. HLA-G is thought to act by suppressing the activity of uterine NK cells, which normally destroy cells that lack the expression of MHC I.

The understanding of mechanisms of immune regulation is largely derived from the study of autoimmune diseases. Many disease-free individuals possess potentially autoreactive T cells. A variety of mechanisms regulate the response of CD4+ T cells so that they don’t react against self antigens. Naïve T helper cells have the potential to become a variety of specialized T cells. There are now four well-recognized possibilities, each with a unique role and ability to cross-regulate. TH1 cells drive cell-mediated immunity by secreting IL-2 and interferon-γ. TH2 cells drive humoral reactions (antibody and B cell) by secreting IL-4. Regulatory T cells are a subtype that suppresses ongoing cellular immune reactions through cell contact. Lastly, there is a newly described proinflammatory population of T cells (TH17) that secrete IL-17. These TH17 cells under normal circumstances are important for the clearance of parasites, bacteria, and fungi, but under pathologic conditions, they appear to play a crucial role in the development of autoimmune disease. One of the hallmarks of T-cell regulation is the ability of these specialized T-cell populations to cross-regulate.


The mother’s immunologic defense system remains intact during pregnancy. While allowing the fetus to grow, the mother must still be able to protect herself and her fetus from infection and antigenically foreign substances. The nonspecific (innate) mechanisms of the immunologic system (including phagocytosis and the inflammatory response) are not affected by pregnancy. The specific (adaptive) mechanisms of the immune response (humoral and cellular) are also not significantly affected. In fact, women with renal transplants do not experience any reduction in serious episodes of acute rejection during pregnancy. No significant change occurs in the leukocyte count. The percentage of B or T lymphocytes is not appreciably altered, nor is there any consistent alteration in their performance during pregnancy. Immunoglobulin levels do not change in pregnancy, and vaccine responses are preserved.

However, pregnant women are at higher risk for severe infection and death from certain pathogens such as viruses (hepatitis, influenza, varicella, cytomegalovirus, polio), bacteria (Listeria, streptococcus, gonorrhea, salmonella, leprosy), and parasites (malaria, coccidioidomycosis) compared with nonpregnant women. The underlying mechanism for this selective immune suppression is not clearly understood.


The major pregnancy-associated immunologic disease process is hemolytic disease of the newborn. Rh factor incompatibility, which is the most important of these conditions, is discussed in Chapter 15.

Hemolytic disease secondary to non-Rh sensitization and the destruction of lymphocytes or platelets secondary to sensitization against specific surface antigens have the same pathogenesis. Fetal cellular antigens leak into the maternal circulation, primarily at birth, and initiate an immune response. The reaction to these foreign antigens (primarily Rh) leads to a humoral response. Initially, only a weak IgM response can be measured. In a subsequent pregnancy, the maternal immune system undergoes a memory response, and highly specific IgG molecules are secreted by memory plasma cells. These antibodies cross the placenta and attach to the fetal Rh-bearing RBCs. The consequence is the sequestration and destruction of fetal RBCs in the fetal spleen, leading occasionally to profound fetal anemia and hydrops.

Although the Rhesus antigen (Rh) is the most common cause of fetal alloimmunization-induced fetal anemia, other antigens are also implicated. The Kell antigen has the additional problem that the maternal IgG against Kell also suppresses erythropoiesis in the fetal bone marrow. ABO incompatibility does not lead to a significant maternal immune response to fetal antigens. Thus, the nature of the antigen is important, but the reason certain antigens are potentially pathogenic is poorly understood.


Fineman J.R., Clyman R., Heymann M.A. Fetal cardiovascular physiology. In: Creasy R.K., Resnick R., Iams J.D., editors. Maternal-Fetal Medicine. Principles and Practice. 5th ed. Philadelphia: WB Saunders; 2004:169-180.

Kahn D.A., Koos B.J. Maternal physiology during pregnancy. In: Decherney A.H., Nathan L., Goodwin T.M., Laufer N., editors. Current Diagnosis and Treatment. Obstetrics & Gynecology. 10th ed. New York: McGraw-Hill; 2007:149-158.

Koos B.J. Breathing and sleep states in the fetus and at birth. In: Marcus C.L., Carroll J.L., Donnelly D.F., Loughlin G.M., editors. Sleep and Breathing in Children. 2nd ed. New York: Informa Healthcare; 2008:1-17.

Terness P., Kallikourdis M., Betz A., et al. Tolerance signaling molecules and pregnancy: IDO, galectins, and the renaissance of regulatory T cells. Am J Reprod Immunol. 2007;58:238.