Chapter 22. Evaluation of the Pregnant Patient

1. Unlike other operations in which the patient is primarily concerned with himself or herself, the pregnant woman is usually concerned for her infant's welfare.

2. The anesthesia provider must be aware of the various physiologic changes of pregnancy and incorporate them into the anesthetic plan. These changes may affect various concurrent diseases if present.

3. The central nervous system effects of pregnancy include a reduced local anesthetic requirement when these agents are administered intrathecally or epidurally.

4. Pregnant patients in the third trimester and symptomatic patients in the first/second trimester should be considered at risk for aspiration during general anesthesia.

5. There is extremely weak evidence that surgery during the first trimester is linked to central nervous system defects in the fetus.

6. Fetal heart rate monitoring is possible during some surgical procedures, but it is not universally applied in the United States.

7. Preterm labor and delivery remain the leading cause of perinatal morbidity and mortality in the United States. Preterm labor is difficult to control with medication, with the most promising being the calcium channel blocking drugs. Magnesium sulfate is frequently used for neuroprotection in the preterm neonate.

8. The etiology of preeclampsia remains to be elucidated, but it is believed to be triggered by a paternal antigen in a susceptible mother.

9. Magnesium sulfate is the most effective medication for the prevention of seizures in those with preeclampsia.

10. Labetalol is the preferred drug for the control of blood pressure in mothers with preeclampsia. Antihypertensive medication does not treat the disease process; rather it is used to prevent intracerebral hemorrhage.

11. The 2 causes of antepartum hemorrhage are placenta previa and placental abruption. With the increase in cesarean deliveries, there is a high risk of placenta accreta in patients with previous cesarean delivery and placenta previa.

12. The perinatal transmission of human immunodeficiency virus (HIV) is low if the viral load is less than 1000 copies/mL and these patients do not require cesarean delivery. If the viral load is greater, cesarean delivery may decrease the risk of perinatal transmission.

13. Aggressive treatment of gestational diabetes mellitus, even mild gestational diabetes, with diet modification, oral agents, and insulin, is associated with improved maternal and neonatal outcome.

One of the most rewarding as well as challenging aspects of anesthesia is caring for the parturient because it involves 2 patients, the mother and the fetus. When formulating an anesthetic plan, the outcomes for both patients must be considered, although usually the ultimate welfare of the mother guides final management decisions. Fortunately, in most circumstances, optimal care of the mother provides good care of the fetus. Both preoperative evaluation and formulation of the anesthetic plan require an understanding of the physiologic changes that accompany pregnancy. These physiologic changes result from alterations in the hormonal milieu (accounts for the changes early in pregnancy) and from the mechanical factors of the enlarging uterus (accounts for most of the changes later in pregnancy).

Cardiovascular System

Pregnancy causes an increase in maternal cardiac output, in response to the increased oxygen consumption associated with the developing fetus. Increases in cardiac output begin during the first trimester, resulting initially from an increase in maternal heart rate (typically 10-20 beats/min).1,2 As the pregnancy progresses, there is an increase in stroke volume caused by a reduction in systemic vascular resistance. Cardiac output increases progressively, beginning as early as 10 weeks of gestation and continuing until it is 40% over baseline during the second trimester.3 The increase results in a 10- to 20-fold increase in uterine blood flow. Despite the increase in cardiac output, blood pressure decreases in early pregnancy, reaching a minimum in midpregnancy, and remains low until around 32 weeks of gestation, and then it returns to baseline levels at term. Although systemic vascular resistance decreases during pregnancy, the central venous and pulmonary artery pressures remain similar to the nonpregnant state.

Maternal blood volume increases by 35% during pregnancy. This increase begins in early pregnancy and continues throughout the gestation. The increase in blood volume results from an increment in both extracellular and intravascular volume. Several theories attempt to explain this increase in blood volume. The "underfill" hypothesis proposes that blood volume expands as a result of systemic vasodilation and an attempt to accommodate the increased intravascular space. In contrast, the "overfill" hypothesis suggests that the increase results from the greater levels of salt-retaining steroids that accompany gestation, with the intravascular space expanding to accommodate the increase in volume.4 Most likely, the increase results from a combination of the 2 theories: there is an initial "underfilling" very early in gestation, and the increase in mineral corticoid levels occurs later in gestation. Whatever the mechanism, the increased blood volume during pregnancy allows parturients to tolerate normal blood loss of delivery (approximately 400 mL during vaginal delivery and 700 mL during cesarean delivery).5 Further, the uterus contracts at delivery, resulting in autotransfusion of approximately 500 mL. Blood volume returns to prepregnancy levels within 7 to 14 days after delivery.

Vascular responsiveness is altered during pregnancy: Normal parturients are less responsive to vasopressors and chronotropic agents.6,7 In animals, both the carotid and uterine arteries responded less to norepinephrine.8 This decreased response may be related to downregulation of α and β receptors. The clinical implications of decreased vascular responsiveness are 2-fold. First, treatment of hypotension may require an increased amount of vasopressor to treat hypotension as compared with that required in the nonpregnant individual. Second, the response to an intravenous test dose of epinephrine will not increase the heart rate to the same extent as in the nonpregnant individual.9

Pregnant patients are particularly susceptible to cardiovascular alterations induced by changes in body position. In the supine position, up to 15% of pregnant patients develop nausea, hypotension, and vomiting (supine hypotension syndrome).10 The supine position allows the gravid uterus to compress the inferior vena cava and aorta. Caval compression results in decreased venous return to the right atrium and hence decreased cardiac output. Anesthetic drugs or techniques that cause venodilation further decrease venous return accompanying caval obstruction. This hypotension decreases uterine perfusion. Uterine blood flow may be further compromised if the uterus compresses the aorta because the uterine artery is a branch of the hypogastric artery, which emerges distal to the level of aortic compression. Studies performed with pregnant women in the lateral position have not noted major decreases in cardiac output.11,12 By tilting the patient to the left, the uterus is displaced off the vena cava and aorta (Fig. 22-1). As such, pregnant women should not lie supine after 20 weeks of gestation, especially if anesthesia is being provided. Whenever the supine position is required, the uterus should be tilted to the left by placing a wedge underneath the right hip, and this principle applies not only at the time of delivery but throughout pregnancy. For example, during cardiopulmonary resuscitation, effective chest compressions may only be accomplished with the parturient in left uterine displacement.13

Physical examination of pregnant women typically reveals a point of maximal cardiac impulse that is displaced cephalad and leftward from its location prepregnancy. Diaphragmatic elevation displaces the heart leftward so that it may appear enlarged on chest radiographs. The electrocardiogram shows left axis deviation. A small pericardial effusion may develop in healthy asymptomatic women during the third trimester of pregnancy.14

Respiratory System

Various changes occur in the maternal airway during gestation. There is vascular engorgement of the airway, resulting in edema of the oral and nasal pharynx, larynx, and trachea.15 When parturients were examined at 12 weeks of gestation and then again at 36 weeks of gestation, the percentage of patients with a Mallampati class IV airway had doubled.16 This airway edema may render intubation of the trachea difficult. This supposition is confirmed by examining maternal deaths from anesthesia in the United States, with approximately half occurring because of failed intubation.17 For oral tracheal intubation, the likelihood of edema in the false cords mandates the use of smaller endotracheal tubes. A 6.0 cuffed endotracheal tube should be ready and available for every case. Exacerbation of these changes may occur in patients with upper respiratory tract infections or preeclampsia. The mucous membranes are also very friable. Manipulation, such as nasal intubation or the insertion of a nasogastric tube, may result in excessive bleeding.

The gravid uterus results in a 4-cm elevation of the diaphragm; the circumference of the thorax is increased 5 to 7 cm. Despite this elevation, there is little change in total lung capacity as the chest expands in both anterior-posterior and transverse diameters to compensate. The diaphragmatic elevation does cause a 20% decrease in functional residual capacity (FRC) at term. This decrease is a result of decreases in both residual volume and expiratory reserve volume. Oxygen consumption increases by 20% due to increased metabolism and increased work of breathing.18 The parturient compensates for this increased oxygen consumption in 2 ways: (1) increased alveolar ventilation, and (2) shifting the oxyhemoglobin dissociation curve to the right, thus facilitating unloading of oxygen at the cellular level. The P50 increases from 26 to 30 mm Hg. This shift is caused by an increase in 2,3-diphosphoglycerate.

The decrease in FRC and increased oxygen consumption make parturients very vulnerable to hypoxia. After complete denitrogenation by breathing 100% oxygen, nonpregnant patients tolerate up to 9 minutes of apnea before oxygen saturation is less than 90%, whereas parturients can only tolerate 2 to 3 minutes. The decrease in FRC allows for the rapid denitrogenation, with 4 deep breaths of 100% oxygen equivalent to 3 minutes of tidal volume breathing of 100% oxygen.19

Minute ventilation increases by 3 to 4 L/min.20 This increase results primarily from increased tidal volume and secondarily from a small increase in respiratory rate. Alveolar ventilation also increases by 70%. Hormone-mediated increases in the neural drive contribute to the hyperventilation of pregnancy. The PaCO2 decreases to 32 mm Hg as a result of increased ventilation and PaO2 increases 5 to 10 mm Hg. The respiratory alkalosis is only partially compensated by a lowering of plasma concentrations of bicarbonate from 26 to 22 mm Hg.

Approximately 60% to 70% of healthy pregnant women have mild, intermittent dyspnea. The cause is not clear but may be related to alterations in chest wall proprioreceptors or progesterone-induced hyperventilation.21 Diaphragmatic elevation may result in increased lung markings, mimicking mild congestive heart failure, on the chest radiographs of healthy pregnant women.

Gastrointestinal Changes

Parturients should be considered to be at risk for pulmonary aspiration due to the physiologic changes of pregnancy. The enlarged gravid uterus displaces the stomach cephalad. This displacement alters the angle of the gastroesophageal junction, decreasing competence of the gastroesophageal sphincter. The uterus also displaces the pylorus upward and posteriorly, resulting in delayed gastric emptying. Elevated concentrations of progesterone decrease gastrointestinal motility and food absorption. These changes facilitate the occurrence of gastric reflux and heartburn in as many as 70% of pregnant women.22 Recent studies question the delayed gastric emptying and increased gastrin concentration. Eleven women had gastric emptying times measured during the first trimester, third trimester, and postpartum.23 There was no difference between first trimester and postpartum. Gastrointestinal transit time was significantly longer in the third trimester. Thus, despite hormonal changes, it seems that gastric emptying is only affected in the third trimester and during labor. This supposition was further confirmed when the gastric contents were aspirated via a nasogastric tube prior to cesarean delivery.24 The stomach contents of 100 term parturients were compared with those from 100 nonpregnant women scheduled for gynecologic surgery. The gastric volume in the pregnant patients was greater than that in the nonpregnant group. Unlike previously believed, serum gastrin levels did not differ between the groups (Fig. 22-2).

These findings affect approaches to the care of the pregnant patient who presents for nonobstetric anesthesia care. Gastric emptying is not delayed until the third trimester. Greater gastric volume occurs but not early in the pregnancy, and serum levels of gastrin are not elevated. From the perspective of pulmonary aspiration of gastric contents, parturients during the first and second trimesters should be treated like nonpregnant individuals. If the patient has symptomatic reflux, rapid sequence induction and intubation should be done. If asymptomatic, rapid sequence induction and intubation are not required. During the third trimester, it may be prudent to perform rapid sequence induction and intubation, although the literature does not strongly support this contention. Clinical judgment is important here, weighing the risks of rapid sequence induction (failed intubation, tachycardia, hypertension) versus the risks of aspiration in an asymptomatic parturient.

Hepatic Changes

There is little or no change in hepatic blood flow during normal pregnancy. Healthy pregnant women frequently have an increase in serum aspartate aminotransferase, lactic dehydrogenase, alkaline phosphatase, or cholesterol. Colloid oncotic pressure decreases as a result of decreased total protein and a decreased albumin-to-globulin ratio. Colloid oncotic pressure decreases further after delivery, regardless of method of delivery or anesthesia administered.25

Pseudocholinesterase activity is decreased 24% before delivery and by 33% on the third postpartum day.25,26 It returns to normal 2 to 6 weeks postpartum. The decreased cholinesterase activity is usually not sufficient to result in clinically relevant prolonged paralysis after a single dose of succinylcholine.

Hematologic Changes

Plasma volume increases 45% (increasing approximately 1000 mL from onset of pregnancy to 36 weeks of gestation), but the red cell mass increases only 20%, leading to the physiologic (dilutional) anemia of pregnancy. Hemoglobin concentration typically decreases by 2.4 g/dL from prepregnancy to 36 weeks of gestation; hematocrit typically decreases 6.5%.27 A maternal hemoglobin concentration less than 11 g/dL is abnormal. The most common cause of anemia during pregnancy is iron deficiency.

Platelet count is usually elevated, but one can see a modest decrease without any other hematologic pathology. It is believed that gestational thrombocytopenia is an exaggerated normal response in which platelets are consumed.28 Pregnancy produces a hypercoagulable state. All coagulation factors, except factors XI and XIII, increase in concentration. Fibrinolytic activity decreases during the third trimester as a result of decreased plasminogen activator concentrations. These changes combined with rapid contraction of the uterus after placental separation help protect women from major hemorrhage. Unfortunately, this hypercoagulability also predisposes pregnant women to thromboembolic complications. Thromboembolism is the leading cause of maternal mortality.29 Parturients are at greatest risk for deep venous thrombosis and pulmonary embolism immediately after delivery.

Endocrine

Mean blood glucose concentration remains within the normal range, although glucose concentration may be lower in some women than those before pregnancy. Parturients become insulin resistant due to human placental lactogen, causing an increased release of insulin.30

Recommendations for weight gain have changed, as documented by the Institute of Medicine. Current guidelines are based on the World Health Organization cutoff points for body mass index. The recommended weight gain during pregnancy is presented in Table 22-1.31 The greatest change is the narrow range of weight gain for obese parturients.

The thyroid gland enlarges 50% to 70% during pregnancy. Concentrations of thyroid-binding globulin and total thyroxine increase, but concentrations of free thyroxine and tri-iodothyronine do not change.

Renal

Renal plasma flow and glomerular filtration rate (GFR) increase during the first trimester to 50% above normal by the fourth month. The increase in renal plasma flow and GFR result in an increased creatinine clearance, with a decreased blood urea nitrogen (BUN) and creatinine.32 BUN decreases 40% to 8 to 9 g/dL; creatinine decreases to 0.4 to 0.5 mg/dL. A BUN of 15 mg/dL, a serum creatinine of 1.0 mg/dL, or a creatinine clearance of 100 mL/min suggest abnormal renal function in pregnant women who are near term. Glucosuria may occur in healthy pregnant women because tubular glucose reabsorption may not keep up with increased GFR. Proteinuria, up to 300 mg/d, is common.

Maternal progesterone, which is a smooth muscle relaxant, causes dilation of renal calyces, pelves, and ureters. After midpregnancy, the enlarged uterus compresses the ureters of the pelvic brim, exacerbating ureteral dilation. Urinary stasis predisposes pregnant women to urinary tract infections.

Dermatologic

Hyperpigmentation is the most common dermatologic change in pregnancy; it occurs in approximately 90% of pregnancies.33 It is usually more evident in women of darker complexion. This hyperpigmentation occurs in the areolar area and also in the clinical appearance of preexisting freckles and nevi. Thickening of scalp hair occurs, and the nails of fingers and toes become softer.

Central Nervous System

The pituitary gland increases in size during pregnancy, but, in contrast, the brain decreases in size.34 The decrease in brain size begins after placental implantation and is maximal at term, reversing slowly after delivery. The precise mechanism and physiologic importance of these changes is not known.

The minimum alveolar concentration for inhaled anesthetics is decreased up to 40% in pregnancy.35 This decrease occurs with all of the inhalation agents. The mechanism is unclear, although it may be related to progesterone (which has sedative activity) and endorphins. A concentration of an inhalation agent that may not produce loss of consciousness in nonpregnant patients may render pregnant women unconscious, placing the parturient at risk for aspiration. For this reason, inhalation agents for labor analgesia are not desirable.

Pregnant women require less total dose of local anesthetic to produce the same level of epidural or spinal block.36 In the epidural space, it may be partly due to epidural vein engorgement, thus decreasing the volume of the epidural space. However, this decreased requirement is seen in the first trimester, well before significant mechanical changes have occurred. Acid base changes in cerebrospinal fluid or hormonal changes of pregnancy may cause increased sensitivity to local anesthetics.37

Reproductive Tract

The uterus weighs 50 to 70 g in the nonpregnant state and increases to 1.0 to 1.5 kg in pregnant women at term. Total uterine blood flow increases from 50 mL/min in nonpregnant women to 700 mL/min in pregnant women at term, representing approximately 10% of the cardiac output. Most of the uterine blood flow perfuses the intervillous space; the remainder of the blood flow perfuses the enlarged myometrium.

The uterine vasculature appears maximally dilated in healthy pregnant women; therefore, uterine blood flow decreases parallel with maternal arterial pressure. There is little or no autoregulation of uterine blood flow during pregnancy. Any factor that decreases venous return and cardiac output will decrease uterine blood flow. Uterine contractions also decrease uterine blood flow. No known drug specifically or directly increases uterine blood flow.

The use of electronic fetal monitoring to assess fetal well-being has become universal, used by both physicians and nurses.38 The American College of Obstetricians and Gynecologists has stated that the goal of fetal surveillance is the prevention of fetal death.39 However, the problem with the current testing methods relates to their predictive value. The negative predictive value is high, meaning that the chance of death within 1 week of a reassuring test is rare. The same does not apply to positive predictive value. A nonreassuring (ie, "positive") test does not imply that the fetus is at risk of death. The obstetrician considers the gestational age of the fetus, other fetal indices, and judgment to arrive at a decision regarding the management of a nonreassuring test.

Monitoring the fetal heart rate is used to determine adequate cerebral oxygenation of the fetus. As the brain modulates the heart, a decrease in fetal heart rate is believed to reflect inadequate fetal cerebral oxygenation. External fetal heart rate monitors use a Doppler detective device with computerized logic to interpret and count the Doppler signals, whereas internal fetal heart rate monitors involve the placement of an electrode on the fetal scalp. The presence of fetal heart tones as well as its rate and rhythm are used to interpret the status of the fetus. The baseline fetal heart rate (FHR) is determined by approximating the mean FHR rounded to increments of 5 beats/min during a 10-minute interval. A normal baseline heart rate is within 110 to 160 beats/min.40 Baseline FHR variability is defined as fluctuations that are irregular in amplitude and frequency. A normal FHR should have moderate variability defined as an amplitude range of 6 to 25 beats/min. Several characteristics of the FHR depend on fetal gestational age. A preterm fetus is expected to have a more rapid rate with little or no beat-to-beat variability.41

Fetal movement is the easiest means to document fetal well-being. The mother perceives fetal movements, which serves as a basis for assessment. A diminution in the perception of fetal movement often precedes fetal death.42 Perception of 10 distinct movements in a period of up to 2 hours is considered reassuring. The first perception of fetal movement occurs around 20 weeks of gestation. As the pregnancy progresses, the movements are felt on a more regular basis.

Heart rate reactivity is thought to be a good indicator of normal fetal autonomic function. Loss of reactivity is associated most commonly with fetal sleep but also may result from any central nervous system depression. The fetal heart rate typically increases in response to fetal movement. This premise serves as the basis for the nonstress test, in which the mother has the FHR monitored for 20 to 40 minutes and also records the time of fetal movement. A reactive (normal) nonstress test is the presence of 2 accelerations (an increase in fetal heart rate by 15 beats/min lasting for at least 15 seconds) in a 20-minute period. A nonreactive nonstress test is one that lacks accelerations over a 40-minute period.43 The ability of the fetus to accelerate its heart rate depends on the age of the fetus. As such, the nonstress test is usually not used prior to 32 weeks of gestation.

The contraction stress test is based on the response of the fetal heart rate to uterine contractions. During a uterine contraction, fetal oxygenation worsens due to uterine artery compression. In the suboptimally oxygenated fetus, this further decrease in oxygenation is not tolerated, leading to the fetal heart rate pattern of late decelerations. A contraction stress test requires 3 contractions in a 10-minute period. Contractions may be produced by oxytocin infusion.44 The contraction stress test is interpreted according to the presence or absence of late fetal heart rate decelerations, which are defined as decelerations that reach the nadir after the peak of the contraction and persist beyond the end of the contraction.45 The presence of late decelerations suggests that the fetus is compromised and should be delivered, usually by cesarean delivery. The results of the contraction stress test are negative if there are no late decelerations. The test is complex and contraindicated in preterm gestations, parturients with bleeding, and parturients with previous uterine surgery. If the test demonstrates no late decelerations, the incidence of fetal death occurring within 1 week of the negative test results is extremely unlikely. The contraction stress test is rarely used today because of ease of the nonstress test and the biophysical profile.

The biophysical profile (Table 22-2) was developed as a method of integrating real-time observations of the fetus directly in its environment. It is used as an assessment tool for both acute and chronic fetal conditions in the antepartum period. The biophysical profile consists of a nonstress test combined with 4 observations made by ultrasonography.46 The biophysical profile is a reflection of a fetus that can accelerate its heart rate, move appropriately, and is surrounded by a normal amount of amniotic fluid. Each of 5 components is assigned a score of either 2 (normal) or 0 (abnormal). A composite score of 8 to 10 is normal, a score of 6 is considered equivocal, and a score of 4 or less is abnormal. The biophysical profile is frequently performed in the labor suite, necessitating a basic understanding by the anesthesiologist. A score of 8 or 10 has the chance for fetal death within 1 week of the test in normal pregnancies as less than 1 in 1000. A biophysical profile of 6 or less is highly predictive of the need for admission to the neonatal intensive care unit.47

Umbilical artery Doppler flow velocimetry is based on the hemodynamic components of vascular impedance and on the observation that flow velocity waveforms in the umbilical artery of normally growing fetuses differ from those of growth-restricted fetuses.48 Various blood vessels have been investigated, including the maternal uterine artery, fetal middle cerebral artery, and umbilical artery. Commonly measured flow velocities include peak systolic-flow velocity and the diastolic flow velocity occurring at the end of the cardiac cycle. The umbilical flow velocity waveform of normally growing fetuses is characterized by high-velocity diastolic flow, whereas with intrauterine growth restriction, there is a diminution of umbilical artery diastolic flow. Commonly measured flow indices are based on peak systolic frequency shift (S) and end diastolic frequency shift (D). The poor predictive value of Doppler velocimetry limits its use to the setting of fetal growth restriction. A S/D ratio of less than 3 is normal, even in the setting of growth restriction, and supports continuing the pregnancy.

Future monitors may include the transabdominal fetal pulse oximeter. The depth of the fetal head from the maternal abdomen is determined on ultrasound; the distance between the optodes needs to be twice the depth of the fetus. Using this device, oxygen saturations between 50% and 74% have been obtained and are similar to those obtained transvaginally.49 The monitor provides measures of both the fetal heart rate and the oxygen saturation of the fetal blood. The value of this monitor is evident because not all cases of decreased fetal heart rate variability are due to decreased oxygenation. The transabdominal fetal pulse oximeter allows for the differentiation between normal fetal sleep and decreased oxygen content.

Preterm birth resulting in a premature neonate is a major health problem in the United States. The overall rate of preterm birth in the United States in 2006 was 12.8%; black (18.4%) and American Indian/Alaska Native (14.1%) women have the highest risk of preterm delivery as compared with white (11.7%), Asian/Pacific Islanders (10.9%), and Hispanic women (12.2%).50 The concern with preterm labor and birth is its outcome: it accounts for approximately 70% of all neonatal deaths and half of long-term neurologic disabilities.51 Almost a fifth of all very preterm infants do not survive the first year of life. Of the smallest infants, disabilities occur in about half of survivors.52 Health care costs resulting from preterm birth are staggering. The financial burden of preterm birth was estimated to be $26 billion in 2005 or approximately $52,000 per preterm infant.53 Table 22-3 lists the risk factors for preterm labor and delivery.

The pathophysiology of preterm labor is multifactorial and not well understood. Preterm birth may result from early activation of the mechanisms of normal term labor or from pathophysiologic mechanisms that trigger labor by distinct means. Poverty, maternal race, limited maternal education, young maternal age, and inadequate prenatal care increase the risk of preterm labor and delivery. Infection also has a prominent role in preterm birth. It has been estimated that 80% of preterm births are associated with chorioamnionitis or organisms in the placental membrane.54 Finally, there is a genetic predisposition. A strong risk factor for preterm birth is a maternal history of preterm birth. However, the association between polymorphisms in certain genes and the risk of preterm birth has been modest. Given the association between preterm labor and bacterial vaginosis, evidence supports its treatment with metronidazole, which decreases the risk of preterm birth by 25% to 75%.55,56 The other major advance in the prediction of preterm birth comes from genomics, the study of gene expression at the messenger RNA level to provide an integrated view of the relationship between the host genome, gene expression, and disease outcome. Several studies have demonstrated a link between nucleotide polymorphisms and preterm birth.57,58

Tocolysis has been a prominent component of efforts to prevent preterm labor and delivery. The premise behind tocolysis is that, by administering medication, uterine contractility will be reduced, thus retaining the fetus in utero. There is increasing consensus among obstetricians that only a minority of patients benefit from tocolysis. Literally hundreds of trials have evaluated medications for the prevention of preterm birth. Acute tocolysis occurs at the initial presentation for preterm labor, which differs from maintenance after successful tocolysis.

The β-mimetic agents, such as terbutaline or ritodrine, have been increasingly falling from favor as tocolytic agents due to the high incidence of maternal adverse effects, such as hypokalemia, hyperglycemia, tachycardia, decreased systemic vascular resistance, and increased risk of pulmonary edema.59 These drugs have been linked to at least 25 maternal deaths from pulmonary edema. The β-mimetics act by binding to β2-adrenergic receptors on the myometrial cell and by increasing the levels of cyclic adenosine monophosphate, resulting in inactivation of the myosin light chain kinase. In a meta-analysis comparing tocolytic agents, only 1 trial found an increase in the length of pregnancy with β-mimetic agents.60 The American College of Obstetricians and Gynecologists concluded, "It would appear, on the basis of currently available evidence, very difficult to support the use of beta-mimetics, which have a high incidence of side effects. Although these drugs do delay delivery in the short term, there is no demonstrable benefit for the newborn."61 However, these drugs are frequently used to relax the uterus during uterine tetany, which is a sustained, prolonged uterine contraction.

The use of magnesium sulphate for the treatment of preterm labor has undergone several revisions. Although used frequently in the United States, its reason for use has changed. A meta-analysis of magnesium sulphate for tocolysis concluded that it does not reduce the frequency of delivery within 48 hours, 7 days, or preterm birth.62 The effect of magnesium sulfate on the fetus and neonate is controversial. Although some data suggested increased adverse neonatal outcomes or fetal death, some evidence suggests a neuroprotective effect.63 In a systematic review of 5 eligible randomized clinical trials, there was a reduced risk of cerebral palsy with the number needed to treat to prevent 1 case of cerebral palsy being 63. For the mother, magnesium sulfate results in blurry vision and maternal weakness. It potentiates the nondepolarizing muscle relaxants, necessitating a major reduction in dose of these medications.

Calcium channel blockers relax the myometrium by inhibiting the reuptake of calcium in the myometrial wall. The most commonly studied calcium channel blocker for tocolysis is nifedipine which may be administered either orally or sublingually. Maternal side effects include headache and rarely hypotension. A Cochrane review found nifedipine to be superior to β-mimetics in general, with a reduction in the number of women giving birth within 7 days and prior to 34 weeks of gestation.64 These drugs should not be used in women with cardiovascular disease or those who are hemodynamically unstable.

Prostaglandins play a role in the initiation of labor, and, as such, prostaglandin synthetase inhibitors may treat preterm labor and prevent preterm birth.65 The most commonly used prostaglandin synthetase inhibitor is indomethacin, which reduces birth before 37 weeks of gestation. Side effects are primarily fetal and include constriction of the ductus arteriosus and fetal oliguria.

Oxytocin receptor antagonists have been developed in response to the observation that labor is accompanied by an increase in oxytocin receptors. Atosiban is the only oxytocin receptor antagonist that has undergone significant study.66 In regard to its effect on preterm labor, the results have been disappointing. It has no benefit over placebo in regard to increasing the length of labor or infant outcome. No oxytocin-receptor antagonists are yet approved by the Food and Drug Administration for use in the United States.

All tocolytic medications have side effects. The decision of which medication to use depends on the desired outcome. All seem to be effective for short-term tocolysis with only indomethacin and nifedipine demonstrating long-term benefits. In an analysis of 19 trials, the probability for an adverse event from a tocolytic agent was 58% for terbutaline, 27% for nifedipine, 22% for magnesium, and 11% for indomethacin.67 Both nifedipine and indomethacin are inexpensive compared with magnesium sulfate and terbutaline, suggesting that these 2 agents are the better choices for a patient in preterm labor. Magnesium sulfate is used not for tocolysis but rather for the prevention of cerebral palsy.

Preeclampsia is a multisystem disorder unique to human pregnancy. It is characterized by an abnormal vascular response to the placenta that is associated with increased systemic vascular resistance, enhanced platelet aggregation, activation of the coagulation system, and endothelial cell dysfunction. Parturients with preeclampsia develop edema, raised blood pressure, and proteinuria. Preeclampsia occurs typically after the 20th week of gestation, except when associated with a hydatidiform mole, but rarely occurs before the 24th week of gestation.68 The American College of Obstetricians and Gynecologists has defined preeclampsia as a blood pressure of at least 140 mm Hg (systolic) or at least 90 mm Hg (diastolic) on at least 2 occasions at least 4 to 6 hours apart.69 Table 22-4 lists the possible abnormal laboratory values in preeclampsia. In general, maternal and perinatal outcomes are good in patients with mild preeclampsia developing after 36 weeks of gestation. If preeclampsia develops before 33 weeks of gestation or in those patients with preexisting medical disorders, the outcome is less favorable.70 Preeclampsia usually abates within 48 hours of termination of the pregnancy. Eclampsia is preeclampsia with central nervous system involvement leading to seizures or grand mal convulsions not related to other cerebral conditions. The diagnosis of severe preeclampsia requires the presence of end-organ disease (Table 22-5).

The incidence of preeclampsia ranges from 2% to 7% of nulliparous patients.71 It is the third leading cause of maternal mortality in the United States and accounts for 20% of maternal deaths.72 Several risk factors for preeclampsia have been identified (Table 22-6). Preeclampsia tends to occur at both extremes of reproductive age. The greatest risk is a first pregnancy. The importance of the father in the etiology of preeclampsia was highlighted in a study examining the Medical Birth Registry of Norway. If a woman becomes pregnant by a man who has already fathered a pregnancy complicated by preeclampsia in a different woman, her risk of developing preeclampsia is increased 1.8 (95% confidence interval [CI], 1.2-2.6).73 There is also a genetic component. Using the same birth registry, the authors demonstrated that daughters of women who had preeclampsia during pregnancy had doubled the risk of developing preeclampsia themselves. The same did not apply to the sons who fathered children.74

During pregnancy, the placenta exerts a significant influence on maternal blood pressure because of the interaction of its various hormones, vasoactive substances, and structure. In a normal pregnancy, peripheral vascular resistance decreases resulting in a lower blood pressure. Maternal blood pressure decreases as early as 7 weeks of gestation. After 28 weeks of gestation, the blood pressure increases, reaching nonpregnant values toward the end of the third trimester. Also, in a normal pregnancy, all fluid compartments expand. In patients with preeclampsia, the opposite occurs, with an increase in blood pressure and a decrease in the fluid compartments.

The cause of preeclampsia is still unknown, but the disease seems to begin at implantation, well before the clinical manifestations allow the diagnosis to be made. The likely culprit in preeclampsia is the placenta. Preeclampsia is seen in patients with molar pregnancies (thus it cannot be a result of the fetus) and also in patients with abdominal pregnancies (thus it cannot be a result of the uterus). Although the etiology remains unknown, current data suggest that various proteins that control angiogenesis are released from the placenta in women who develop preeclampsia. In a susceptible individual, various angiogenic proteins respond differently from the usual. These proteins are important for normal placental development. Because of this varied response, an imbalance in concentrations occurs, causing abnormal placentation. Angiogenic proteins promote the development of a healthy vasculature; antiangiogenic proteins do the opposite. At 20 to 24 weeks, sFlt-1 (soluble fms-like tyrosine kinase 1) was greater in those women who subsequently developed preeclampsia.75 sFlt-1 is an antiangiogenic factor. It binds placental growth factor (PlGF) and vascular endothelial growth factor (VEGF), 2 factors that promote angiogenesis. The link of sFlt-1 to preeclampsia has been well established. Serum concentrations of this factor are increased in preeclampsia and correlate with the severity of the disease. sFlt-1 does not have a direct effect. It exerts its effect by binding to the angiogenic factors PlGF and VEGF. Although abnormal placentation occurs with the onset of the pregnancy, sFlt-1 levels do not change until the second trimester. Samples analyzed at 10 to 14 weeks of gestation (at least 10 weeks before the onset of disease) were not predictive of preeclampsia. The levels of sFlt-1 in the first trimester in preeclamptic parturients do not differ from those in normal parturients.76 As such, using sFlt-1 and PlGF is not useful early in the pregnancy. The predictive value of these blood tests did not manifest until the second trimester. Another antiangiogenic factor, endoglin, was examined; it is highly expressed on cell membranes of the vascular endothelium and syncytiotrophoblasts of the placenta.77 Placental endoglin is upregulated in preeclampsia, releasing soluble endoglin into the maternal circulation. It is an antiangiogenic protein that inhibits VEGF in the vasculature. Unlike sFlt-1, which increases 5 weeks before the onset of disease, circulating soluble endoglin levels increase 2 to 3 months before the onset of disease. Beginning as early as 17 weeks of gestation, levels of soluble endoglin were higher in women who developed preeclampsia. The combination of measuring both endoglin and sFlt-1 should improve the predictive ability of such tests.78

If preeclampsia is a state of antiangiogenesis, one would predict that the subsequent health of the mother should be affected. Heart disease, an antiangiogenic state, should be increased in women with preeclampsia. Using a meta-analysis involving 3,488,160 women, of which 198,252 had preeclampsia, the relative risks of hypertension (3.70; 95% CI, 2.70-5.05), ischemic heart disease (2.16; 95% CI, 1.86-2.52), and for stroke (1.81; 95% CI, 1.45-2.27) were all increased.79 It seems that the occurrence of preeclampsia has significant implications for the mother's future health, despite the apparent resolution of the acute disease process. A review of systems for women should include a history of preeclampsia.

Anticonvulsants were introduced for women with preeclampsia in the belief that they would prevent eclampsia and improve outcome. Predicting who is at risk for eclampsia is difficult, and only 1% to 2% of those with preeclampsia develop eclampsia. However, prophylactic anticonvulsants are used in all parturients with preeclampsia. A variety of anticonvulsants have been proposed and evaluated, but the most widely used drug is magnesium sulfate. Other agents used include diazepam and phenytoin. Evidence continues to support the use of magnesium sulfate for the prevention of seizures, and it appears to be superior to other agents. The largest study that examined magnesium sulfate was a multicenter trial that included centers from the United Kingdom, South America, and South Africa.80 A total of 10,110 patients were randomized to receive magnesium sulfate (5055 patients) or placebo (5055 patients). There was a greater incidence of side effects in the magnesium sulfate group such as flushing, hot flashes, and blurry vision. The magnesium sulfate group also had a 58% lower risk of eclampsia; this lower risk did not affect maternal mortality. Magnesium does not prevent the progression of the disease to severe preeclampsia. In 222 women with mild preeclampsia, 109 patients received intravenous magnesium. There was no difference in the groups in the progression to severe disease.81 Magnesium sulfate does relax the uterus. The use of magnesium sulfate in parturients with preeclampsia does not slow labor, but there is a greater need for oxytocin.82

It was hoped that low-dose aspirin would be beneficial in preventing preeclampsia. Low-dose aspirin reduces thromboxane production by selective inhibition of platelet thromboxane production without affecting prostacyclin production by the vascular endothelium. The largest study randomized 2539 women to receive either low-dose aspirin (60 mg) or placebo.83 The use of small-dose aspirin did not affect the incidence of preeclampsia. A meta-analysis of aspirin studies identified 14 trials that met specific criteria. Aspirin was only effective in women with known risk factors. It was not beneficial for the general population.84

Treatment of acute hypertension is intended to prevent potential cerebrovascular complications (the most common cause of morbidity in patients with preeclampsia). Antihypertensive medications prevent cerebrovascular problems; they do not alter the course of the disease. Intravenous hydralazine and intravenous labetalol are the most commonly used medications. Originally, it was believed that hydralazine was the drug of choice because it preserves uterine blood flow. In a meta-analysis comparing hydralazine to labetalol, hydralazine was associated with more maternal hypotension, more cesarean deliveries, more adverse effects on the fetal heart rate, and lower 1-minute Apgar scores.85 It was also associated with less maternal bradycardia as compared with labetalol. Labetalol is used more frequently for the treatment of hypertension of preeclampsia.

In a review of hypertension in pregnant women, Lindheimer and Katz concluded, "Epidural block should be avoided, since in preeclampsia it is associated with sudden falls of blood pressure and on occasion with vascular collapse."86 The literature does not support this stance.87,88 Studies have demonstrated the stability of maternal cardiac output after administration of epidural anesthesia in patients with severe preeclampsia. Also, spinal anesthesia is not contraindicated in severe preeclamptic patients.89 Although it does result in a higher incidence of hypotension, these episodes are not more severe than during epidural anesthesia and are also easily treated. In regard to epidural anesthesia, it provides excellent analgesia during labor and allows patients to remain awake and alert during vaginal and cesarean delivery. Epidural anesthesia has at least 3 specific advantages in parturients with preeclampsia who require cesarean delivery. It reduces circulating concentrations of catecholamines in laboring women, facilitating control of blood pressure and improving intervillous blood flow.90,91 It also allows for better control of systemic and pulmonary arterial pressures.92 Finally, epidural anesthesia does not require laryngoscopy and intubation, which might be hazardous or difficult because of pharyngeal edema.93

Platelet turnover and function is often altered in preeclampsia. Normal platelets are disk-shaped cells, 2 to 4 μm in diameter, that lack a nucleus. Platelets are formed in the bone marrow from giant cells called megakaryocytes. At any moment, approximately 80% of the platelets are circulating and 20% are localized in the spleen. Platelets participate in coagulation factor reactions leading to thrombin formation by providing a lipoprotein surface on which coagulation enzymes and substrates interact. The bone marrow does not contain a reserve of platelets. If circulating platelets are rapidly destroyed or lost, thrombocytopenia persists for several days until enough platelets are formed to correct it.

Pitkin and Witte measured platelet counts every 4 weeks in 23 pregnant women and found that platelet counts dropped from 322,000/μL in the first trimester to 278,000/μL in the third trimester.94 This decrease is caused by increased destruction.95 Preeclampsia is accompanied by endothelial injury, leading to increased platelet activation with consumption in the microvasculature. Platelet consumption is a late finding in preeclampsia, but increased platelet turnover may be an early marker of preeclampsia. Thrombocytopenia is the most common hematologic abnormality in patients with preeclampsia. Its incidence depends on the severity of the disease and the presence or absence of abruption placenta. A platelet count less than 150,000/μL was found in 50% of the women and a count of less than 100,000/μL in 36% of women with preeclampsia.96 Roberts et al retrospectively studied 292 patients with low platelets.97 Patients were divided into 2 groups: those with platelet counts less than 50,000/μL and those with platelet counts of 50,000-100,000/μL. Bleeding was only likely if the platelet count was less than 40,000/μL. There is no therapy for the thrombocytopenia of preeclampsia other than delivery. Epidural analgesia has been successfully placed in individuals with platelet counts less than 100,000/μL if there is no other existing coagulopathy.

The 2 major causes of antepartum hemorrhage are placenta previa and placental abruption. Hemorrhage is the second leading cause of maternal mortality in the United States, and its incidence is expected to increase as the number of cesarean deliveries increase.97 Hemorrhage can be quite severe because uterine blood flow at term is 500 to 700 mL/min.

Placenta previa is present when the placenta overlies the cervical os. Placenta previa varies in degree and may be complete (in 37% of cases, the placenta completely covers the internal cervical os), partial (in 27% the placenta partially covers the os), or low lying (in 46% the placenta is in the lower segment but does not reach the internal os). Parturients with either complete or partial placenta previa are at an increased risk of antepartum bleeding, are more likely to have placenta accreta, and are more likely to require hysterectomy and to have postoperative complications than women with low-lying placenta previa.98 Whereas women with complete or partial placenta previa require cesarean delivery, not all parturients with low-lying placenta previa require cesarean delivery. With transvaginal ultrasonography, the location of the placenta may be determined. If the placental edge is greater than 10 mm from the cervical os, the parturient may safely undergo vaginal delivery without increased risk of hemorrhage.99 Risk factors for placenta previa include uterine fibroids and a history of multiple pregnancies (Table 22-7). Placenta previa is more likely to be affected by conditions existing prior to pregnancy, such as maternal age, race, parity, and previous cesarean delivery.100 The main sign of placenta previa is painless vaginal bleeding. The bleeding usually stops spontaneously but may recur suddenly at any time. Recurrent bleeding is usually more severe. If the diagnosis of placenta previa is suspected, ultrasonography is performed. The accuracy of this technique is 95% and has replaced the double setup technique in which a cervical examination is performed in the operating room with preparation to do an immediate cesarean delivery.101,102 Before term gestation, the patient with placenta previa is usually managed conservatively with bed rest if the patient is not actively bleeding. If bleeding does occur, emergency cesarean delivery is required. Occasionally, the placenta may implant directly to the myometrium giving rise to 3 situations: placenta accreta (the placenta attaches to the myometrium), placenta increta (the placenta invades the myometrium but does not penetrate it fully), or placenta percreta (the placenta penetrates the entire thickness of the myometrium and may attach to other structures in the pelvis). In the general obstetric population, the incidence of placenta accreta is 1 in 2500. In patients with placenta previa and no prior cesarean delivery, the incidence is 3%. The risk of placenta accreta in patients with placenta previa and who have had a prior cesarean delivery increases with each prior cesarean delivery. With 1 prior uterine incision, the incidence of placenta accreta has been reported to be 11%; 2 prior uterine incisions, 40%; and 3 prior uterine incisions, 61%.102 In patients with placenta previa and prior cesarean delivery, the anesthesia provider should be prepared for massive hemorrhage. Large-bore intravenous access and the immediate availability of blood products are mandatory.

Placental abruption refers to separation of the placenta after 20 weeks of gestation but before the birth of the fetus. The incidence ranges from 0.2% to 2.4%, depending on the population studied. From 1979 to 2001, the rate of placental abruption increased 92% among black women and 15% among white women.103,104 A high infant mortality is associated with placental abruption. Unlike risk factors for placenta previa, placental abruption is more likely to be affected by conditions occurring during pregnancy, such as maternal cigarette smoking, alcohol consumption, and prenatal care. Risk factors for placental abruption are presented in Table 22-8. The primary etiology of placental abruption is unknown. It begins with hemorrhage into the deciduas basalis, causing a split in the deciduas and subsequent hematoma formation. This expanding hematoma causes additional separation, compression, and ultimately destruction of the adjacent placental tissue. The classic signs of placental abruption are vaginal bleeding, abdominal pain, uterine tenderness, and contractions. Bleeding is either through the vagina or concealed in the uterus. If the blood is concealed, underestimating blood loss is common. Unlike placenta previa, abruption may not be ruled out by a negative ultrasound examination, as ultrasound evidence of hemorrhage is present in only 50% of patients.105 Anesthetic management of the parturient with placental abruption requires volume resuscitation. The anesthesia provider must also be prepared to manage a coagulopathic patient because placental abruption is associated with disseminated intravascular coagulation.

For surgery during pregnancy, the anesthesia provider must consider possible fetal effects of the maternal disease process. The most common cause of the nonobstetric acute abdomen is appendicitis, occurring in 1 in 500 pregnancies and accounting for approximately 25% of surgeries during pregnancy.107 The incidence of acute cholecystitis requiring surgery is 1 in 1600 pregnancies. Approximately 3% of parturients have stones in their gallbladder. For surgery during pregnancy, protection of the mother is paramount, but other goals of anesthetic management include maintenance of uterine blood flow and fetal oxygenation, avoidance of teratogenic drugs, and prevention of preterm labor.

Perhaps the greatest concern of pregnant women is whether anesthetics or anesthesia may increase the risk of congenital anomalies. Shnider and Webster were the first to study this possibility when they evaluated the medical records of 9073 women who delivered infants between July 1959 and August 1964.108 Of these women, 147 (1.6%) had surgery during pregnancy. There was no increased incidence of congenital anomalies in the surgical group, but the authors noted that most of these patients received anesthetics during the second or third trimester (after the period of organogenesis, which is the first trimester). Brodsky et al mailed questionnaires to 287 women who had surgery during pregnancy.109 Among the women, 187 had surgery during the first trimester. There was no major increase in congenital anomalies in infants born to women who had surgery during pregnancy compared with a control group of pregnant women who did not have surgery. Duncan et al, used health insurance data from 1971 to 1978.110 They matched 2565 women undergoing surgery during pregnancy to those of similar height and weight who did not. There was no difference in the rate of congenital anomalies. There was an increased risk of spontaneous abortion in those undergoing surgery with general anesthesia in the first or second trimester, especially in those undergoing gynecologic surgery. Mazze and Kallen obtained data from 3 Swedish health care registries to evaluate the risk of adverse reproductive outcomes after anesthesia administration and surgery in pregnant women between 1973 and 1981.111 They identified 2252 women who underwent surgery during the first trimester. Among these women, 65% received general anesthesia; 35% received regional anesthesia or local infiltration. There was no increase in congenital anomalies among the parturients who had surgery.112 Another approach to this question is to examine anomalies and determine if there is a link to anesthesia or surgery. Of the 20,830 pregnant women who had a child with a congenital anomaly, 31 patients had surgery with anesthesia. This fraction did not differ from the 35,727 women who had infants without congenital anomalies and 73 of them having surgery during pregnancy. There was no higher rate of surgery or anesthesia in any of the congenital anomaly groups.

The lack of link between anesthesia and surgery during pregnancy and congenital anomalies has not been universal. Kallen and Mazze reexamined their database and noted 6 infants with neural tube defects (an overall incidence in the general population is 1 in 1000).113 Of these 6 infants, the mothers of 5 of the infants had surgery during gestational weeks 4 to 5, the period of neural tube formation. This group was not the only one to hypothesize a possible link. Infants born with central nervous system defects in Atlanta between 1968 and 1980 were matched to controls by race, birth hospital, and period of birth.114 Of the 694 mothers of infants with central nervous system defects, 12 reported first-trimester anesthesia exposure (34 of 2984 control mothers reported such exposure). There was an increased risk for hydrocephalus (odds ratio [OR]: 9.6; 95% CI, 3.8-24.6). The strongest association was for hydrocephalus and eye defects (OR: 39.6; 95% CI, 7.5-209.2). These data are influenced by a rare event and by small sample size. In the past, anesthesia providers informed patients that there was a theoretically risk that exposure to anesthesia in the first trimester increased the risk of teratogenesis. These 2 studies raise the possibility of such an existence. When discussing the risk of anesthesia with pregnant women scheduled for surgery, the anesthesia care provider should remind the patient that there is a 3% baseline incidence of fetal anomalies among all pregnant women regardless of exposure.

The use of fetal monitoring during maternal surgery is a debated subject. After 18 to 20 weeks of gestation, the fetal heart rate may be monitored. The argument in favor of intraoperative fetal monitoring is that it may improve fetal outcome. Changes in the fetal heart rate may signal fetal compromise allowing the anesthesia care provider to take steps to improve uteroplacental perfusion and fetal oxygenation. These may include increasing left uterine displacement, higher inspired concentration of oxygen, adjustment of maternal ventilation, augmentation of maternal circulating blood volume, or pharmacologic management of hypotension. Despite these benefits, the use of fetal monitoring during surgery is not universally applied. Hospitals in the United States were surveyed regarding monitoring.115 Of the 184 respondents, 60% routinely used fetal monitors; 40% did not. The American College of Obstetricians and Gynecologists recognized that there are no data regarding monitoring during surgery and stated

It is important for physicians to obtain obstetric consultation before performing nonobstetric surgery because obstetricians are uniquely qualified to discuss aspects of maternal physiology and anatomy that may affect intraoperative maternal-fetal well-being. The decision to use fetal monitoring should be individualized and if used, may be based on gestational age, type of surgery, and facilities available. Ultimately, each case warrants a team approach for optimal safety of the woman and her baby.116

Cardiac Disease

The cardiovascular changes of pregnancy place a major stress on the maternal heart and circulatory system. The changes may even result in death of the mother if she has cardiovascular disease. In an attempt to quantify this risk, 562 pregnant women with cardiac disease (either congenital or acquired) were followed prospectively in 13 Canadian hospitals.117 The study followed these women through pregnancy and for 6 months after delivery. Neonatal morbidity, preterm labor, and small for gestational age births occurred in 20% of these patients. Maternal complications included pulmonary edema, stroke, and tachyarrhythmia. The most significant predictor of morbidity and mortality was the presence of left ventricular dysfunction prior to the pregnancy, with an OR of 11 (95% CI, 4-34). The best predictor of maternal outcome is the New York Heart Association functional class (Table 22-9). In a study of 482 pregnancies, cardiovascular morbidity was less in mothers with class I as compared with others.118 The higher the functional class, the greater the incidence of morbidity and mortality.119 Neonatal complications were seen in 20% and were associated with poor functional class, left heart obstruction, or smoking.

Valvular lesions in pregnant patients are decreasing due to the decrease in rheumatic heart disease. In general, the parturient with valvular incompetence does better than the parturient with stenotic lesions. The reduced systemic vascular resistance of pregnancy improves forward flow, thus limiting the effects of regurgitation. In contrast, stenotic valvular lesions create a fixed impediment to the required increase in cardiac output of pregnancy, precipitating heart failure. Mitral stenosis is the most common valvular lesion encountered in the parturient.120 There is an increased incidence of atrial fibrillation, leading to heart failure. Balloon mitral commissurotomy has been described during pregnancy and is a therapeutic option.121 Aortic stenosis is infrequently encountered and is usually due to a congenitally abnormal valve. Hypertrophic cardiomyopathy is a genetically transmitted cardiac disease characterized by left ventricular hypertrophy and reduced left ventricular size and compliance. Reduction in preload and afterload results in an increase in the outflow gradient and in a reduction in left ventricular filling. One hundred parturients with hypertrophic cardiomyopathy were compared with the general population.122 There were 2 pregnancy-related deaths, both in women with severe disease. If the patient were in a stable condition, the progression of symptoms, atrial fibrillation, and syncope were uncommon.

Congenital heart disease is increasingly prevalent in women of childbearing age due to the advances in diagnosis and treatment of these conditions. Approximately 25% of cardiac disease in the parturient is due to congenital heart disease.123 In pregnant patients with congenital heart disease, both cardiac and neonatal complications are increased. Right subpulmonary ventricular systolic dysfunction and severe pulmonary regurgitation were predictors of an adverse fetal outcome. Congenital heart disease may be divided into 2 groups: those with acyanotic lesions and those with cyanotic lesions. Patients with cyanotic lesions developed significantly more congestive heart failure and deteriorated more often than those with acyanotic lesions. Women with obstructive lesions had a higher incidence of pregnancy-induced hypertension, whereas in the cyanotic group, the mean birthweight was approximately 1 kg less than that the acyanotic group. Acyanotic conditions include atrial or ventricular septal defects or persistent ductus arteriosus with small or moderate left to right shunts.124 Cyanotic lesions include tetralogy of Fallot and transposition of the great vessels with subsequent Fontan procedure. In cyanotic lesions, the increase in cardiac output and decrease in systemic vascular resistance lead to greater right-to-left shunting, worsening the cyanosis and hypoxemia.

Peripartum cardiomyopathy is a rare form of heart failure with no identifiable cause that occurs within the last month of pregnancy or within 5 months after delivery. The incidence is 1 in 4000. Risk factors include advanced maternal age, multiparity, multiple gestation, black population, obesity, and preeclampsia. There is a high morbidity and mortality with approximately 20% of those developing the condition dying.125 The remainder of those affected recovers partially or completely. The largest series of women who had the disorder and a subsequent pregnancy involved 44 women.126 Twenty-eight pregnancies occurred in women who recovered completely; 16 pregnancies occurred in women who had persistent left ventricular dysfunction. All deaths occurred in the patients with persistent left ventricular dysfunction and they were also twice as likely to present with congestive heart failure. These patients also had a greater frequency of premature deliveries.

Acute myocardial infarction during pregnancy is rare, with an incidence of 1 in 10,000 pregnancies. However, the incidence is increasing, most likely reflecting the trend toward an older maternal age.127 Mortality is approximately 40%, with the greatest risk occurring if the infarction occurs late in the pregnancy. Cardiac troponin is unaffected by pregnancy, allowing for the diagnosis of ischemic infarction.

Pulmonary hypertension is defined as a mean pulmonary artery pressure greater than 25 mm Hg associated with pulmonary capillary wedge pressure less than 12 mm Hg. Primary pulmonary hypertension is a rare disease that affects young women of childbearing age. Pulmonary hypertension may also be secondary to other causes such as congenital heart disease, thromboembolic disease, and connective tissue disorders. The most common cause of pulmonary hypertension in the parturient involves the late consequences of a large cardiac shunt (eg, Eisenmenger syndrome). Pulmonary hypertension is tolerated poorly during pregnancy due to insufficient adaptation of the right heart to increases in cardiac output and a poorly compliant pulmonary vasculature. In a series of 14 parturients with severe pulmonary hypertension (4 primary and 10 secondary), 2 patients died before delivery and 3 died postpartum.128 It was not possible to determine whether parturients with primary pulmonary hypertension fared better than patients with secondary pulmonary hypertension.

Asthma

Asthma is a disease of chronic airway inflammation with acute episodes of bronchospasm. Asthma severity is classified according to its clinical features (Table 22-10). Approximately 7% of pregnant women are affected by asthma.129 A number of physiologic changes of pregnancy affect the course of asthma in the parturient. Airway closure that results from decreased functional residual capacity during tidal breathing might lead to exacerbation of the disease.130 To determine the effect of pregnancy and stage of pregnancy on asthma, 6 electronic databases were searched for prospective studies of asthmatic women during pregnancy. This review demonstrated that asthma improves in 70% of parturients and worsens in the remaining 30%. This improvement peaked in the second trimester and reverted to baseline after delivery. Peak expiratory flow rate declines significantly throughout gestation, returning to normal at 6 weeks postpartum.131

Parturients with asthma are at a higher risk of developing complications during pregnancy. Parturients with poorly controlled asthma had a doubling of risk for spontaneous preterm labor.133 If the asthma is controlled, there is no increased risk of preterm delivery, although there is an increased risk of stillbirth for the male fetus.134 Frequency of asthmatic symptoms during pregnancy was strongly associated with preeclampsia. Fetal hypoxia has been suggested as the mechanism for the association between asthma and preeclampsia as well as between asthma and intrauterine growth retardation.135 Parturients with asthma have a greater risk of developing chorioamnionitis, most likely due to premature rupture of membranes coupled with immunosuppressive therapy of asthma.136

The treatment of asthma combines allergen avoidance, smoking cessation, pharmacotherapy, and education. The value of smoking cessation was reinforced in a study of 2210 women. Active smoking in women with asthma had a strong association with increased asthma symptoms and fetal growth abnormalities; passive smoke exposure did not exacerbate the disease process.137 The management of pregnant women with asthma should follow the same guidelines as for nonpregnant patients, with inhaled steroids as the first line of therapy.138 Inhaled steroids, even in high doses, is not associated with congenital malformations in the neonate.139 Parturients are not treated with systemic glucocorticoids due to its teratogenetic effect (the risk of orofacial cleft increases by 3).140 Treatment with inhaled steroids as compared with theophylline results in a greater improvement in FEV1 in the parturient.141 The mainstay for treatment of an acute attack is inhaled albuterol. Albuterol is a relatively selective β2-adrenergic bronchodilator. It has been shown to have no effect on maternal and fetal circulations.142

Back Pain

Back pain during pregnancy is a frequent problem. A cohort of 200 consecutive women was followed throughout their pregnancy.143 Of these participants, none reported back pain at the beginning of the study. However, at 12 weeks of gestation, 19% of the study population complained of back pain. The incidence increased to 47% at 24 weeks of gestation and peaked at 49% at 36 weeks of gestation. After delivery, the prevalence of back pain declined to 9.4%. Despite a relatively high prevalence, only 32% of women with low back pain during pregnancy report this problem to their physician, and only 25% of providers recommend a specific therapy.144 If patients receive osteopathic manipulation during the third trimester, the progression of back pain may be halted.145

The etiology of the back pain is multimodal. One popular theory is that the exaggerated lumbar lordosis to compensate for the enlarging uterus places significant mechanical strain on the lower back. There is also a hormonal component. Relaxin is a polypeptide hormone of ovarian origin of the insulin-like growth factor family. Relaxin is associated with remodeling of collagen fibers and pelvic connective tissue. The primary source of circulating relaxin is the corpus luteum. The placenta is also a major source of relaxin. When serum levels of relaxin were measured in parturients with and without back pain, these levels best explained differences in back pain.146

Patients who develop low back pain during pregnancy may avoid subsequent pregnancy to prevent recurrence of the back pain. Women with low back pain during pregnancy have an extremely high risk for experiencing a new episode during a subsequent pregnancy.147

Most patients with low back pain during pregnancy respond to activity and postural modification. Exercise to increase the strength of the abdominal and back muscles is helpful. Scheduled rest periods with elevation of the feet to flex the hips and decrease the lumbar lordosis help to relieve muscle spasm and pain.148

Diabetes Mellitus

Diabetes mellitus complicates 4% to 7% of pregnancies in the United States, although the incidence is expected to increase with the increase in frequency of gestational diabetes mellitus.149 Gestational diabetes mellitus is defined as glucose intolerance that first occurs during pregnancy. The reason for the increase is the increase in maternal weight. A large weight gain during the first trimester of pregnancy, especially in overweight or obese women, increases a woman's risk of gestational diabetes mellitus.150 Type 1 diabetes results from primary failure of endogenous insulin production, whereas type 2 diabetes represents a relative insulin deficiency. The precise etiology of type 1 diabetes is unknown; it appears that to develop type 1 diabetes mellitus, one must have genetic susceptibility and an environmental insult, which leads to an autoimmune attack against the insulin-producing pancreatic islet β cells. (See Chapter 13 for a detailed discussion of endocrine disease.)

During normal pregnancy, the parturient develops an insulin resistance that begins near midpregnancy and progresses through the third trimester. The insulin resistance appears to be a result from a combination of increased maternal adiposity and of hormones excreted by the placenta, such as human placental lactogen and progesterone. The fact that insulin resistance rapidly abates following delivery (important to remember when managing diabetic parturients in the labor suite) suggests that the major contributor to this insulin resistance is placental hormones.

Glucose control is important for both maternal and fetal well-being during pregnancy. Studies have consistently shown a significant positive correlation between ambient serum glucose concentration during organogenesis and the incidence of spontaneous abortion.151 When glycosylated hemoglobin levels were less than 8.5%, the fetal malformation rate (cardiac defects, sacral agenesis, renal agenesis, polycystic kidneys, anencephaly, meningomyelocele) was 3.4%, but when the glycosylated hemoglobin level was greater than 9.5%, the rate of fetal malformations approached 22%.152 Glucose crosses the placenta; fetal levels reflect maternal levels. Insulin does not cross the placenta. Maternal hyperglycemia produces fetal hyperglycemia causing stimulation of the fetal β cells and fetal hyperinsulinemia. Insulin is the major fetal growth hormone and produces excessive fetal growth, especially in the fat. The fetus of a poorly controlled diabetic mother is likely to be large, especially in the shoulders and chest, increasing the risk of shoulder dystocia at delivery. Given this risk, there is a higher incidence of cesarean delivery as compared with the general population, with 45% to 73% of parturients with type 1 diabetes mellitus having cesarean delivery. A weight gain of 15 kg and suspected macrosomia are risk factors associated for cesarean delivery without labor.153 Increased insulin levels in the fetus delays fetal lung maturity. Diabetic parturients also have an increased risk of developing preeclampsia.154 In 1949, White proposed a system for classifying diabetes in obstetric patients155 (Table 22-11). The White classification is still used today as a means of conveying to others the severity of the disease process.

The management of diabetes in pregnancy includes a careful combination of diet, exercise, and insulin. An aggressive approach to mild gestational diabetes including dietary intervention and self-monitoring of blood glucose with insulin if necessary was associated with a significant reduction in fetal overgrowth, shoulder dystocia, cesarean delivery, and preeclampsia.156 Any abnormality of blood glucose should be aggressively managed to improve outcome for mother and fetus. Fasting blood glucose should be maintained near 90 mg/dL and the 1-hour postprandial glucose below 140 mg/dL. During the night, glucose levels should not fall below 60 mg/dL. Glycosylated hemoglobin levels should be no higher than 6%.157

Gestational diabetes mellitus is defined as glucose tolerance during pregnancy. Risk factors include advanced maternal age, family history of diabetes, and obesity.158,159 Long-term follow-up of patients with gestational diabetes indicates that most progress to diabetes after pregnancy.160 Very few have diabetes soon after delivery; the incidence appears to increase 10 or more years after delivery. The focus of treatment antepartum is to return glucose levels to normal. After pregnancy, the main focus of clinical care should be reducing the risk of diabetes and treating it if it develops.

Much debate has focused on the role of oral hypoglycemic agents for the management of diabetes during pregnancy. These drugs are routinely used in nonpregnant adults with type 2 diabetes. They are also effective in parturients with gestational diabetes. The most commonly used agent is glyburide, the second-generation sulfonylurea. It does not cross the placenta, thus avoiding possible teratogenicity. It has an onset of action of 4 hours and duration of 10 hours. In a study of 404 women, glyburide was similar to insulin in regard to glucose control, with a lower incidence of hypoglycemia in the glyburide group.161 When compared with glyburide in 149 parturients with gestational diabetes, the failure rate of metformin was 2.1 times higher than the failure rate of glyburide.162

Thyroid Disease

Thyroid disease is more common in women, with a female-to-male ratio of 12:1. (See Chapter 13 for details regarding endocrine disease.) Thyroid autoimmunity is by far the most frequent cause of hypothyroidism in women of reproductive age.163 The prevalence of hypothyroidism in the general population of reproductive age is 2%. The incidence of hyperthyroidism in pregnant women is estimated at 0.2%, with the most common cause being Graves disease (autoimmune-induced thyroxine overproduction). During pregnancy, normal thyroid activity undergoes significant changes, including a 2- to 3-fold increase in thyroxine-binding globulin concentrations, increased serum thyroglobulin, and increased renal iodide clearance. In addition, human chorionic gonadotropin has mild thyroid stimulating activity.164

Hypothyroidism in women may cause infertility. This cause is not surprising because thyroid hormones have direct effects on granulose cells, luteal cells, and oocytes. With mild hypothyroidism, pregnancy may occur, but the resulting pregnancies are often associated with abortion, stillbirth, or prematurity.165 For the fetus, maternal thyroid hormones are transferred across the placenta. This transfer is important in early gestation because the fetal thyroid gland becomes operational only after midgestation. During pregnancy, the parturient requires a greater iodine intake. Before pregnancy, a woman requires approximately 100 μg/d, which increases to 200 to 250 μg/d to maintain free thyroxine during pregnancy.166 This higher iodine requirement is a result of the increase in thyroxine production by the mother, a transfer of iodine to the fetus, and an increased renal iodine clearance. In the United States, approximately 11% of parturients do not receive sufficient iodine. Maternal hypothyroidism is associated with neurologic disorders in the newborn because thyroid hormonal levels are important for fetal brain development.167 In 2007, the Endocrine Society Clinical Practice Guideline on the Management of Thyroid Dysfunction During Pregnancy and Postpartum was published. In a study of 4562 parturients, universal screening reduced the incidence of adverse obstetrical and neonatal outcomes in a low-risk population.168

Hyperthyroidism during pregnancy is less common than hypothyroidism. Hyperthyroidism may lead to miscarriage, neonatal death, preterm delivery, and intrauterine growth retardation. Two pregnancy-specific conditions, hyperemesis gravidarum and trophoblastic disease, may cause hyperthyroidism in the mother.169,170 Hyperemesis gravidarum is characterized by severe vomiting that begins around 6 weeks of gestation and usually resolves by 20 weeks of gestation. The etiology of transient hyperthyroidism in hyperemesis gravidarum is unclear. In trophoblastic disease, there is an increase in human chorionic gonadotropin, which increases thyroid function. In select cases, propylthiouracil may be required.

Human Immunodeficiency Virus

Acquired immunodeficiency syndrome (AIDS) was first reported in women in 1981 with the percentage of AIDS cases in women increasing thereafter. For AIDS among children younger than 3 years, perinatal transmission accounts for 92% of the cases.171 Of the women with HIV, 31% were of childbearing age. Pregnancy does not affect HIV or disease progression; levels of virus are not increased by pregnancy.172 As such, the challenge with HIV in pregnant patients is the prevention of transmission of HIV to the child.

The first major discovery in the prevention of transmission of the virus to the neonate was the demonstration that treatment of the mother with zidovudine during pregnancy and labor and of the neonate during the first 6 weeks of life reduces the transmission rate from 25% to 8%.173 The introduction of this practice had a dramatic effect on perinatal transmission.174 However, the use of a single agent is considered obsolete, and patients now receive a combination therapy. Combination therapy consists of 2 nucleoside reverse transcriptase inhibitors and 1 nonnucleoside reverse transcriptase inhibitor or protease inhibitor. Combination antiretroviral therapy has been proven effective in reducing maternal HIV-1 RNA levels to less than 500 copies/mL, which minimizes the risk of perinatal transmission from a baseline of 26% to 1% to 2%.175 Given that an effective therapy exists, the Centers for Disease Control and the American College of Obstetricians and Gynecologists have recommended that all pregnant women be screened for HIV. HIV-negative infants born to HIV-positive mothers frequently exhibit a range of immunologic abnormalities.176

Early studies suggested an association between vaginal delivery and fetal transmission of the virus, with a decreased likelihood of transmission with cesarean delivery.177,178 In 1999, the International Perinatal HIV Group published a meta-analysis that included only those studies of at least 100 mother-child pairs and defined elective cesarean delivery as that performed before the onset of labor.179 The number of patients in the final analysis was 7800 mother-child pairs. The study concluded that the likelihood of transmission from the mother to the child decreased by approximately 50% with elective cesarean delivery as compared with other modes of delivery. The addition of antiretroviral therapy during prenatal, intrapartum, and neonatal periods reduced the risk further. The problem with this analysis was that most studies were performed prior to the use of highly active antiretroviral therapy and also without any data concerning maternal viral load. Given these concerns, the American College of Obstetricians and Gynecologists published a committee opinion, which may be summarized as follows:

1. In patients with HIV who did not take antiretroviral therapy, the risk of vertical transmission is 25%.

2. The addition of zidovudine decreases the risk to 5% to 8%.

3. If a cesarean delivery is performed, the risk is approximately 2%.

4. A similar risk of 2% or less is seen among women with viral loads of less than 1000 copies/mL. Women whose viral loads are greater than 1000 copies/mL should be counseled regarding the potential benefit of scheduled cesarean delivery.

5. Viral load should be measured to assist the obstetrician in planning the route of delivery.180

Cocaine Abuse

The consequences of acute and chronic cocaine abuse must be considered in the obstetric population, due to the prevalence of abuse in women of childbearing age. Overall approaches to the perioperative management of substance abuse are provided in Chapter 24; this brief review focuses on the special concerns that affect the parturient and her child. Cocaine is derived from the Erythroxylum coca plant. It is taken intravenously, intranasally, or orally. Crack is a form of cocaine that is smoked. Cocaine produces sympathetic stimulation by blocking the presynaptic reuptake of norepinephrine, dopamine, and serotonin. This pathophysiology explains the hypertension and tachycardia frequently seen following acute use. In fact, it is difficult to differentiate preeclampsia from acute cocaine toxicity based on physiologic variables.181 Urine testing for protein differentiates between the 2.182 Almost 90% of drug-abusing women are of reproductive age, with an estimated 4.6 million female users of cocaine in the United States and 750,000 drug-exposed births annually.183 In addition to its cardiovascular effects, maternal ingestion of cocaine is associated with premature labor, placental abruption, nonreassuring fetal heart rate, and uterine rupture. Cocaine rapidly crosses the placenta resulting in a high incidence of fetal anomalies. In children who were exposed in utero to cocaine, performance on tasks that assess sustained attention and behavioral self-regulation are compromised.184 Evidence indicates that this deficiency is a result of disruption of neuronal pathways associated with attention and behavioral self-regulation. In the mother, cocaine is associated with thrombocytopenia. Propanol is contraindicated in patients with hypertension from acute cocaine use due to unopposed α-adrenergic stimulation. Labetalol or hydralazine is acceptable. The American Heart Association recommends nitroglycerine and benzodiazepines as first-line agents for patients experiencing cocaine-related myocardial ischemia. For hypotension, ephedrine may not be effective due to a depletion of catecholamines. Phenylephrine would be effective.185

Unlike other patients, the evaluation of the parturient involves 2 patients, the mother and fetus. Optimal care of the mother provides the best care for the fetus. The parturient undergoes various physiologic changes that vary depending on gestation. These changes must be considered when devising an anesthetic plan. These changes also must be considered when the parturient has an underlying disease. These diseases and conditions typically worsen during pregnancy, ultimately resulting in deterioration of the mother's condition. These changes also have implications in the pharmacologic management. Preterm labor does not entail increased risk to the mother; rather, it carries significant risk to the fetus. However, the use of tocolytic agents places the mother at risk. Magnesium sulfate is typically administered during preterm labor, not necessarily for tocolysis but rather for the neuroprotective effects in the neonate. Magnesium sulfate also remains the cornerstone for seizure prophylaxis in patients with preeclampsia. Magnesium sulfate potentiates both nondepolarizing (significantly) and depolarizing muscle relaxants. Although one of the most challenging aspects of anesthesia is the care of the parturient, it also is one of the most rewarding.

• Berkman ND, Thorp JM Jr, Lohr KN, et al. Tocolytic treatment for the management of preterm labor: a review of the evidence. Am J Obstet Gynecol. 2002;186:587.
• Bonnin M, Mercier FJ, Sitbon O, et al. Severe pulmonary hypertension during pregnancy. Anesthesiology. 2005;102:1133.
• Cohen SE, Andes LC, Carvalho B. Assessment of knowledge regarding cardiopulmonary resuscitation of pregnant women. Int J Obstet Anesth. 2008;17:20.
• Czeizel AE, Pataki T, Rockenbauer M. Reproductive outcome after exposure to surgery under anesthesia during pregnancy. Arch Gynecol Obstet. 1998;261:193.
• Doyle LW, Crowther CA, Middleton P, Marret S. Antenatal magnesium sulfate and neurologic outcome in preterm infants: a systematic review. Obstet Gynecol. 2009;113:1327.
• Hong JY, Park JW, Oh JI. Comparison of preoperative gastric contents and serum gastrin concentrations in pregnant and nonpregnant women. J Clin Anesth. 2005;17:451.
• Jensen D, Webb KA, O'Donnell DE. Chemical and mechanical adaptations of the respiratory system at rest and during exercise in human pregnancy. Appl Physiol Nutr Metab. 2007;32:1239.
• Kain ZN, Mayes LC, Ferris CA, Pakes J, Schottenfeld R. Cocaine-abusing parturients undergoing cesarean section: a cohort study. Anesthesiology. 1996;85:1028.
• Kwon H, Belanger K, Bracken B. Effect of pregnancy and stage of pregnancy on asthma severity: a systematic review. Am J Obstet Gynecol. 2004;190:1201.
• Landon MB, Spong CY, Thom E, et al. A multicenter, randomized trial of treatment for mild gestational diabetes. N Engl J Med. 2009;361:1339.
• Macones GA, Hankins GDV, Spong CY, Hauth J, Moore T. The 2008 National Institute of Child Health and Human Development Workshop Report on Electronic Fetal Monitoring: Update on Definitions, Interpretation, and Research Guidelines. J Obstet Gynecol Neonatal Nurs. 2008;37:510.
• Macones GA, Parry S, Elsousy M, et al. A polymorphism in the promoter region of TNF and bacterial vaginosis: preliminary evidence of gene-environment interaction in the etiology of spontaneous preterm birth. Am J Obstet Gynecol. 2004;190:1504.
• Negro R, Schwartz A, Gismondi R, Tinelli A, Mangieri T, Stagnaro-Green A. Universal screening versus case finding for detection and treatment of thyroid hormonal dysfunction during pregnancy. J Clin Endocrinol Metab. 2010;95:1699.
• Oatridge A, Holdcroft A, Saeed N, et al. Change in brain size during and after pregnancy: study in healthy women and women with preeclampsia. AJNR Am J Neuroradiol. 2002;23:19.
• Pilkington S, Carli F, Dakin MJ, et al. Increase in Mallampati score during pregnancy. Br J Anaesth. 1995;74:638.
• Silver RM, Landon MB, Rouse DJ, et al. Maternal morbidity associated with multiple repeat cesarean deliveries. Obstet Gynecol. 2006;107:1226.
• Siu SC, Sermer M, Colam JM, et al. Prospective multicenter study of pregnancy outcomes in women with heart disease. Circulation. 2001;104:515.
• Strathearn L, Mayes LC. Cocaine addiction in mothers: potential effects on maternal care and infant development. Ann NY Acad Sci. 2010;1187:172.
• Wang SM, Dexinno P, Maranets I, et al. Low back pain during pregnancy: prevalence, risk factors, and outcomes. Obstet Gynecol. 2004;104:65.
• Whittaker PG, Macphail S, Lind T. Serial hematologic changes and pregnancy outcome. Obstet Gynecol. 1996;88:33.