Cardiovascular Changes

Changes in the cardiovascular system occur throughout gestation and include (1) anatomic changes, (2) an increase in intravascular volumes, (3) an increase in cardiac output, (4) a decrease in vascular resistance, and (5) the presence of supine hypotension. Table 62.1 and the following sections detail these changes.

Aortocaval Compression

Aortocaval compression by the gravid uterus as a result of supine positioning is associated with a decrease in systemic blood pressure. Although the inferior vena cava is compressed in nearly all term parturients, supine hypotension syndrome (also known as aortocaval compression syndrome) is experienced by only 8% to 10% of women. Supine hypotension syndrome is defined as a decrease in mean arterial pressure of more than 15 mm Hg with an increase in heart rate of more than 20 beats/min and is often associated with diaphoresis, nausea, vomiting, and changes in mentation. At term, the inferior vena cava can be almost completely occluded in the supine position with the return of blood from the lower extremities through the epidural, azygos, and vertebral veins that become engorged ( Fig. 62.1 A ). Also, significant aortoiliac artery compression occurs in 15% to 20% of pregnant women. Inferior vena caval compression in the supine position causes a decrease in both stroke volume and cardiac output of 10% to 20% in comparison to the upright position (see Fig. 62.1 B ), and may exacerbate venous stasis in the legs and thereby result in ankle edema, varices, and increased risk for lower extremity deep venous thrombosis.

Aortocaval compression.

(A) Cross-sectional views of aortocaval compression from the gravid uterus in the supine position with loss of compression in the lateral position. (B) Alterations in heart rate, stroke volume, and cardiac output for both supine and lateral positioning with increasing gestation of pregnancy. IVC, Inferior vena cava.

Reprinted with permission from Bonica JJ, ed. Obstetric Analgesia and Anesthesia . Amsterdam: World Federation of Societies of Anaesthesiologists; 1980.

Most pregnant women have compensatory adaptations that reduce supine hypotension symptoms despite aortocaval compression. One compensatory response is a reflexive increase in peripheral sympathetic nervous system activity. This increase in sympathetic activity results in increased systemic vascular resistance and permits arterial blood pressure to be maintained despite the reduced cardiac output. Consequently, the reduced sympathetic tone from neuraxial or general anesthetic techniques impairs the compensatory increase in vascular resistance and exacerbates the impact of hypotension from supine positioning.

Therefore in general, supine positioning is avoided during use of neuraxial techniques for labor analgesia and cesarean deliveries. Reducing the compression of the inferior vena cava and abdominal aorta with left tilt may mitigate the degree of hypotension and help maintain uterine and fetal blood flow. This is accomplished by positioning the patient laterally or by elevating the right hip 10 to 15 cm (with a historical goal of 15 degree left-tilt) with a blanket, wedge, or table tilt.

The practice of left uterine displacement has been challenged recently. In a magnetic resonance imaging (MRI) study of healthy pregnant volunteers, the volume of the inferior vena cava did not differ significantly between the supine position and the 15 degree left-tilt position but when the patients were tilted to the 30 degree left-tilt position, the inferior vena cava volume did increase. Additionally, healthy women undergoing elective cesarean delivery under spinal anesthesia and a phenylephrine infusion were randomized to supine or 15 degree left-tilt position and no difference was found on neonatal acid-base status; however, the supine patients had lower cardiac output and required more phenylephrine. Further studies are needed to investigate who benefits from left uterine displacement and the amount required to achieve the greatest benefit without hindering the surgical procedure. In the meantime, left uterine displacement should continue to be utilized during induction of neuraxial analgesia/anesthesia and during episodes of maternal hypotension or fetal compromise.

Respiratory System Changes

Pregnancy results in significant alterations in (1) the upper airway, (2) lung volumes and ventilation, and (3) O ₂ consumption and metabolic rate ( Table 62.2 ).

Gastrointestinal Changes

After midgestation, the induction of general anesthesia places pregnant women at increased risk for regurgitation, aspiration of gastric contents, and development of acid pneumonitis. The stomach and pylorus are moved cephalad by the gravid uterus, which repositions the intraabdominal portion of the esophagus intrathoracically and decreases the competence of the lower esophageal sphincter muscle. Higher progesterone and estrogen levels of pregnancy further reduce lower esophageal sphincter tone. Gastrin, secreted by the placenta, increases gastric hydrogen ion secretion and lowers the gastric pH in pregnant women. These changes in combination with the increased gastric pressure from the enlarged uterus increase the risk for acid reflux in pregnancy. Maternal gastric reflux with subsequent esophagitis (heartburn) is common in pregnant women and increases with increased gestational age. Gastric emptying is not prolonged in pregnancy. Conversely, gastric emptying is decreased with the onset of labor, pain, anxiety, or administration of opioids. Increased gastric contents can further increase the risk for aspiration. Epidural analgesia using local anesthetics alone does not further delay gastric emptying; however, epidural boluses of fentanyl can cause gastric emptying delay.

All women in labor are considered to have a full stomach and an increased risk for pulmonary aspiration of gastric contents during induction of anesthesia. To reduce this risk, a nonparticulate antacid, a rapid sequence induction of anesthesia technique including cricoid pressure, and endotracheal intubation are considered routine parts of general anesthesia in a pregnant woman past midgestation.

Hepatic and Biliary Changes

Blood flow to the liver does not change significantly with pregnancy. The markers of liver function, including aspartate aminotransferase (AST), alanine aminotransferase (ALT), and bilirubin, increase to the upper limits of normal with pregnancy. Alkaline phosphatase levels more than double secondary to placental production. Plasma protein concentrations are reduced during pregnancy, and the decreased serum albumin levels can result in elevated free blood levels of highly protein-bound drugs. Plasma cholinesterase (pseudocholinesterase) activity is decreased approximately 25% to 30% from the 10th week of gestation up to 6 weeks postpartum. Although neuromuscular transmission should be analyzed before extubation, the clinical consequence of the reduced cholinesterase activity is unlikely to be associated with marked prolongation of the neuromuscular block resulting from succinylcholine. The risk for gallbladder disease is increased during pregnancy with incomplete gallbladder emptying and changes in bile composition. Acute cholecystitis is the second most common cause of acute abdomen in pregnancy and occurs between 1 in 1600 to 1 in 10,000 pregnancies.

Renal Changes

Renal blood flow and the glomerular filtration rate (GFR) increase during pregnancy. Renal blood flow rises 60% to 80% by midpregnancy and in the third trimester is 50% greater than nonpregnant values. GFR is increased 50% above baseline by the third month of pregnancy and remains elevated until 3 months postpartum. Therefore the clearance of creatinine, urea, and uric acid are increased in pregnancy, and the upper laboratory limits for blood urea nitrogen and serum creatinine concentrations are decreased approximately 50% in pregnant women. Levels of urine protein and glucose are commonly increased as a result of decreased renal tubular resorption capacity. The upper limits of normal in pregnancy in a 24-hour urine collection are 300 mg protein.

Hematologic Changes

As previously discussed, blood volume increases during pregnancy. At term, the plasma volume has increased approximately 50% above prepregnancy values and the red cell volume has increased only approximately 25%. The greater increase in plasma volume creates a physiologic anemia of pregnancy with a hemoglobin value normally around 11.6 g/dL. Hemoglobin values less than this at any time during pregnancy are concerning for anemia. Overall oxygen delivery is not reduced by the normal physiologic anemia of pregnancy because of the subsequent increase in cardiac output. The additional intravascular fluid volume of approximately 1000 to 1500 mL at term helps compensate for the estimated blood loss of 300 to 500 mL typically associated with vaginal delivery and the estimated blood loss of 800 to 1000 mL that accompanies a standard cesarean delivery. After delivery, contraction of the evacuated uterus creates an autotransfusion of blood often in excess of 500 mL that offsets the blood loss from delivery.

Leukocytosis is common in pregnancy and is unrelated to infection. Leukocytosis is defined as a white blood cell (WBC) count greater than 10,000 WBCs/mm ³ of blood. In pregnancy, the normal range can extend to 13,000 WBCs/mm3. WBC count may rise in labor with the degree of increase related to the duration of elapsed labor. The WBC count may decrease over the first week postpartum but may take weeks or months to return to nonpregnant values.

Neurologic Changes

Pregnant patients are considered more sensitive to both inhaled and local anesthetics. They have a reduced minimum alveolar concentration (MAC) for inhaled anesthetics. The MAC of a volatile anesthetic is reduced by 40% in animals and 28% in humans during the first trimester of pregnancy. However, it appears the decrease in MAC (i.e., immobility in response to volatile anesthetics among 50% of patients) occurs at the level of the spinal cord based on an electroencephalographic study suggesting that anesthetic effects of sevoflurane on the brain are similar in the pregnant and nonpregnant state. The underlying mechanism of reduced MAC in pregnancy remains unclear; it is likely multifactorial, and many postulate progesterone may have a role.

Pregnant women are more sensitive to local anesthetics and neuraxial anesthetic requirements are decreased by 40% by term. At term, the epidural veins are distended and the volume of epidural fat increases, which decreases the size of the epidural space and volume of cerebrospinal fluid (CSF) in the subarachnoid space. Although the decreased volume of these spaces facilitates the spread of local anesthetics, the local anesthetic dose requirement is decreased for neuraxial block as early as the first trimester, before significant aortocaval compression or other mechanical- or pressure-related changes occur. Consequently, the increased nerve sensitivity and decrease in local anesthetic dose requirements are likely biochemical in origin.

Uterine Blood Flow

An understanding of uteroplacental blood flow is critical for appropriate clinical care. Uterine blood flow increases progressively during pregnancy from about 100 mL/min in the nonpregnant state to 700 to 900 mL/min (∼10% of cardiac output) at term gestation. Approximately 80% of the uterine blood flow perfuses the intervillous space (placenta) and the remainder perfuses the myometrium. Uterine blood flow has minimal autoregulation, and the vasculature remains essentially fully dilated during pregnancy. Uterine and placental blood flow depend upon maternal cardiac output and are directly related to uterine perfusion pressure and inversely related to uterine vascular resistance. Decreased perfusion pressure can result from maternal hypotension secondary to multiple causes including hypovolemia from blood loss or dehydration, decreased systemic resistance from general or neuraxial anesthesia, or aortocaval compression. Increased uterine venous pressure from aortocaval compression, frequent or prolonged uterine contractions, or prolonged abdominal musculature contraction with bearing down (Valsalva) during second-stage pushing can decrease uterine perfusion. Additionally, extreme hypocapnia (Pa CO ₂ < 20 mm Hg) occasionally associated with hyperventilation secondary to severe labor pain can reduce uterine blood flow with resultant fetal hypoxemia and acidosis. Neuraxial blockade does not alter uterine blood flow as long as maternal hypotension is avoided but decreases in maternal blood pressure during neuraxial or general anesthesia should be immediately corrected.

Endogenous maternal catecholamines and exogenous vasopressors may cause increasing uterine arterial resistance and decreasing uterine blood flow depending on the class and amount given. In a pregnant ewe model, use of α-adrenergic vasopressors—methoxamine and metaraminol—increased uterine vascular resistance and decreased uterine blood flow, whereas administration of ephedrine did not reduce uterine blood flow despite drug-induced increases in maternal arterial blood pressure. As a result, ephedrine was previously considered the vasopressor of choice for the treatment of hypotension caused by the administration of neuraxial anesthesia to pregnant women. In complete contrast, more recent human trials demonstrate the use of phenylephrine (α-adrenergic agonist) for prophylaxis or treatment of neuraxial-induced hypotension is not only effective in preventing hypotension, but also is associated with less fetal acidosis and base deficit than the use of ephedrine. Other methods to reduce maternal hypotension with induction of regional or general anesthesia are discussed in the section on anesthesia for cesarean delivery.

Placental Exchange

Fetal Circulation and Physiology

Fetal blood volume increases throughout gestation. Approximately onethird of the fetal-placental blood volume is contained within the placenta. During the second and third trimester, the fetal blood volume is estimated to be approximately 120 to 160 mL/kg of fetal body weight. Thus the total blood volume of a term normal fetus is approximately 0.5 L. Although fetal liver function is not yet mature, coagulation factors are synthesized independent of the maternal circulation. The serum concentrations of these factors increase with gestational age and do not cross the placenta. However, fetal clot formation in response to tissue injury is decreased in comparison to that in adults.

The anatomy of the fetal circulation helps decrease fetal exposure to potentially high concentrations of drugs in umbilical venous blood. Approximately 75% of umbilical venous blood initially passes through the fetal liver, which may result in significant drug metabolism before the drug reaches the fetal heart and brain (first-pass metabolism). Fetal and neonatal enzymatic drug metabolism activities are lower than those of adults, but most drugs can be metabolized. In addition, drugs entering the fetal inferior vena cava via the ductus venosus, thus bypassing the portal circulation and the liver, are initially diluted by drug-free blood returning from the fetal lower extremities and pelvic viscera. These anatomic characteristics of the fetal circulation add to the complexity of maternal-fetal pharmacokinetics.

Labor and Fetal Monitoring

Intrapartum fetal monitoring was created to evaluate fetal well-being and detect fetal distress earlier in labor to allow intervention prior to permanent fetal injury. Electronic fetal monitoring (EFM) combines interpretation of fetal heart rate (FHR) monitoring and uterine contraction monitoring. FHR monitoring was developed in the 1960s and its use has been increasing since. There is high interobserver and intraobserver variability of FHR tracing interpretation. A meta-analysis comparing EFM to intermittent FHR auscultation noted the use of EFM reduced the risk for neonatal seizures (relative risk [RR]: 0.50), but not the risks for perinatal mortality or cerebral palsy. The use of this monitoring has been shown to increase the rate of both operative and cesarean deliveries.

The nomenclature, interpretation, and management principles for FHR monitoring were updated in 2009 by the American Congress of Obstetricians and Gynecologists (ACOG). These current guidelines are detailed later, and related terminology is presented in Box 62.1 . An understanding of the specific uterine contraction and FHR monitoring terminology as well as the clinical implications is critical for optimal communication during emergent situations among anesthesiologists, obstetricians, midwives, and labor nurses.

Contraction Monitoring

Uterine contractions can be monitored externally with a tocodynamometer or internally with an intrauterine pressure transducer. External monitors only allow determination of contraction frequency, whereas internal monitors also allow quantitative measurement of intrauterine pressure. The Montevideo unit is traditionally used by obstetricians to assess the adequacy of uterine contractions. The Montevideo unit is defined as the intensity of contractions (in millimeters of mercury, as measured with an intrauterine pressure catheter) multiplied by the number of contractions that occur in 10 minutes.

Uterine contractions are quantified over a 10-minute window that is averaged over a 30-minute window with guidelines provided by the ACOG. Normal contractions are defined as five or fewer contractions in 10 minutes, averaged over a 30-minute window. Tachysystole is defined as uterine activity greater than five contractions in 10 minutes, averaged over a 30-minute window. Tachysystole applies to both spontaneous and augmented labor and should always be qualified as to the presence or absence of associated FHR decelerations. Treatment of tachysystole during labor may differ depending on the clinical situation but may include sublingual or intravenous nitroglycerin to briefly relax the uterus, as well as the use of β ₂ -adrenergic drugs such as terbutaline.

Fetal Heart Rate Tracing

FHR monitoring is most commonly accomplished with a surface Doppler ultrasound transducer (external monitoring), but it may be necessary to apply a fetal scalp electrode to obtain accurate continuous FHR monitoring (internal monitoring). For internal monitoring, a peak or threshold voltage of the fetal R wave from the scalp electrode is used to measure FHR. Of note, a fetal scalp electrode can be placed only if the cervix is minimally dilated and the membranes are ruptured. The FHR pattern changes in response to fetal asphyxia from activation of peripheral and central chemoreceptors and baroreceptors. It also shows changes as a result of various fetal brain metabolic changes that occur with asphyxia. These changes in the FHR produce specific patterns and characteristics that provide an evaluation of the fetal state.

The FHR tracing is used as a nonspecific reflection of fetal acidosis. It should be interpreted over a time course in relation to the clinical context and other known maternal and fetal comorbidities, because multiple factors other than fetal acidosis can influence the FHR tracing. Box 62.1 defines FHR baseline, variability, and accelerations. A normal baseline FHR ranges from 110 to 160 bpm. FHR variability are fluctuations in the baseline FHR that are irregular in frequency and amplitude. Normal FHR variability predicts early neonatal health and a fetal central nervous system that is normally interacting with the fetal heart. Accelerations are abrupt changes in the FHR above baseline and are defined by gestational age of the fetus.

Fig. 62.3 details FHR tracing deceleration characteristics. Late decelerations are a result of uteroplacental insufficiency causing relative fetal brain hypoxia during a contraction. The resulting sympathetic outflow elevates the fetal blood pressure and activates the fetal baroreceptors and an associated slowing in the FHR. A second type of late deceleration is from myocardial depression in the presence of increasing hypoxia. Therefore late decelerations are considered worrisome. On the other hand, early decelerations are considered benign and tend to mirror the uterine contraction and are believed to be in response to vagal stimuli, which are often the result of fetal head compression. Variable decelerations are associated with umbilical cord compression. A sinusoidal FHR pattern is associated with fetal anemia and is considered ominous. In general, minimal-to-undetectable FHR variability in the presence of variable or late decelerations is associated with fetal acidosis. Prolonged decelerations (<70 beats/min for >60 seconds) are associated with fetal acidemia and are extremely ominous, particularly with the absence of variability.

Fetal heart rate deceleration.

(A) Early deceleration: visually apparent, usually symmetric gradual decrease and return of the fetal heart rate (FHR) with the nadir of the deceleration occurring at the same time as the peak of the contraction. (B) Variable decelerations: visually apparent, abrupt decrease in FHR ≥15 seconds and <2 minutes. The relationship of the onset of the deceleration compared to the contraction is variable as is the depth and duration. (C) Late FHR decelerations: visually apparent, usually symmetric gradual decrease and return of the FHR during which the nadir of the deceleration occurs after the peak of the contraction.

Fetal Heart Rate Categories

A three-tiered FHR category classification system is currently recommended for fetal assessment with the specific criteria for each category outlined in Box 62.2 . This system evaluates the fetus for the given moment of the assessment. The fetal condition may move back and forth among the categories over time. Specific terminology used for categorization is defined in Box 62.1 .

Category I FHR tracings are considered normal and are predictive of a normal fetal acid-base state at the time of observation, and no specific management is required.

Category II FHR tracings are considered indeterminate and include all tracings not in categories I or III. Category II tracings are not predictive of abnormal fetal acid-base status and require continued monitoring and reevaluation with consideration for the entire clinical picture. In some cases, additional tests may be obtained or intrauterine resuscitative measures taken to improve the fetal condition.

Category III FHR tracings are considered abnormal and are associated with an abnormal fetal acid-base state at the time of observation. These tracings require prompt patient evaluation and efforts to improve the fetal condition. These interventions may include intrauterine resuscitation with change in maternal position, discontinuation of labor augmentation, treatment of maternal hypotension with fluids and/or vasopressor administration, use of supplemental O ₂ , and/or administration of a tocolytic agent such as terbutaline. If the FHR tracing does not improve, expeditious delivery should occur which may involve an assisted vaginal (forceps or vacuum) delivery or a cesarean delivery.

Nonpharmacologic Labor Pain Management

Many patients prefer to use nonpharmacologic methods of pain management during all or part of labor. Acupuncture can be effective in treating postoperative pain after cesarean delivery, but it is not as effective for analgesia during labor. A systematic review and meta-analysis of acupuncture for pain relief in labor involving 10 randomized controlled trials ( n = 2038) found that acupuncture was not superior to sham acupuncture (superficial needling lateral to an actual acupuncture point) at 1 and 2 hours. Unfortunately, most of the trials were not properly blinded, increasing the likelihood of bias.

Several trials have found a reduction in pain and anxiety during the first stage of labor with the use of massage. A Cochrane review on massage in labor identified seven randomized trials of massage, six of which were judged to have low or unclear risk for bias. During the first stage of labor, pain was reduced in the massage group by −0.98 (confidence interval [CI], 1.17-0.47) on a 10-point pain scale. No difference was found in the use of pharmacologic pain relief between groups or in pain reported during the second and third stages of labor. Of note, one study of 60 women found that massage decreased anxiety.

Hypnosis has been used both as a relaxation technique and for management of pain during labor. When hypnosis was compared with standard care, no evidence was found that pain was less with the use of hypnosis, nor was evidence found for a difference in satisfaction with pain relief. A Cochrane review of 9 trials randomizing 2954 women found women in the hypnosis group were less likely to use pharmacologic pain relief compared to those in the control groups. There were no clear differences for satisfaction, spontaneous vaginal birth, and postpartum depression between the two groups.

Other nonpharmacologic techniques include the breathing techniques described by Lamaze, the LeBoyer technique, the Bradley method, transcutaneous nerve stimulation, hydrotherapy, presence of a support person, intradermal water injections, and biofeedback. A retrospective national survey of women’s childbearing experiences in the United States found that although neuraxial methods of pain relief were considered the most helpful and effective, nonpharmacologic methods of hydrotherapy and massage were rated more or equally helpful in relieving pain compared with the use of intravenous opioids. Although many nonpharmacologic techniques seem to reduce the perception of labor pain, most published studies lack the rigorous scientific methodology for useful comparison of these techniques to pharmacologic methods.

Considerations for Pharmacologic Treatment of Pain in Labor

Preprocedural assessment by an anesthesia provider should be performed for any candidate for neuraxial labor analgesia. Clinical assessment should be considered for all patients admitted to a labor and delivery floor not only to discuss labor analgesia options prior to excruciating pain, but also to assess the patient for comorbid conditions that could complicate labor, obstetric procedures, or anesthesia. The obstetric anesthesia team should be prepared to care for all admitted patients in the event of an obstetric emergency. In otherwise healthy women, laboratory testing is not required during a routine preprocedural obstetric assessment.

Although any laboring woman has the potential to require cesarean delivery, labor takes many hours and requires adequate nutrition and hydration. While balancing these two considerations, the ASA has recommended that moderate amounts of clear liquids be allowed during the administration of neuraxial analgesia and throughout labor. A period of abstention from solids before the placement of neuraxial analgesia is not required. However, the ASA does recommend the ingestion of solid foods be avoided in laboring patients.

Systemic Medications

Opioids can be used for labor analgesia. They are inexpensive, widely available, and can be administered intramuscularly without the need for intravenous access. Although there are differences among opioids, they all cross the placenta and can have fetal effects including dose-related respiratory depression and decreased FHR variability.

Although meperidine was once the most commonly used long-acting opioid in obstetric practice, it is the most likely to result in side effects. Meperidine is typically administered intravenously in doses of up to 50 mg or intramuscularly in doses ranging from 50 to 100 mg. Maternal half-life of meperidine is 2.5 to 3 hours whereas the half-life for its active metabolite normeperidine is 13 to 23 hours. The half-life of both is up to three times longer in the fetus and newborn. Normeperidine can accumulate with repeated doses and can be neurotoxic. With increased dosing and shorter intervals between doses and delivery, risk to the newborn is increased, including lower Apgar scores and prolonged time to sustained neonatal respiration.

Morphine is rarely used for labor pain. Like meperidine it has an active metabolite (morphine-6-glucuronide) with a half-life that is longer in neonates than in adults, and it produces significant maternal sedation. Obstetricians may use intramuscular morphine for analgesia, sedation, and rest. This produces analgesia with an onset of 10 to 20 minutes and is used most commonly in latent labor. Maternal side effects may include respiratory depression and histamine release resulting in pruritus and rash.

Mixed agonist-antagonist opioid analgesics such as nalbuphine and butorphanol are utilized to treat labor pain. Nalbuphine has similar analgesic potency as morphine. It is given either intravenously, intramuscularly, or by subcutaneous injection at doses of 10 to 20 mg every 4 to 6 hours. Butorphanol is five times as potent as morphine and 40 times more potent than meperidine. A dose of 1 to 2 mg intravenously or intramuscularly is commonly used for labor analgesia. Both medications are often well tolerated by the parturient.

Fentanyl and, more recently, remifentanil have become popular systemic opioid analgesics over the past two decades. Fentanyl is a synthetic opioid that is highly lipid-soluble and has a short duration and no active metabolites. When given in small IV doses of 50 to 100 μg/h, no significant differences are seen in neonatal Apgar scores and respiratory effort compared with those in newborns of mothers not receiving fentanyl. Fentanyl is also commonly used in patient-controlled analgesia (PCA) during labor. Common doses utilized for fentanyl PCA include a bolus dose of 10 to 25 μg with a lockout interval of 5 to 12 minutes. High doses of systemic fentanyl, especially immediately prior to birth, could result in neonatal depression.

Remifentanil PCA may offer superior pain relief and lesser fetal effects than other intravenous opioid analgesics but its analgesic effects are inferior to epidural labor analgesia and it requires careful maternal oxygenation and ventilation monitoring. The metabolism of remifentanil depends completely on tissue and plasma esterases, which are fully mature in the fetus. Further, it is more rapidly metabolized in the placenta (by placental esterases) than in the maternal plasma and thus the fetal-to-maternal ratio is small. In the pregnant ewe model, the maternal-to-fetal ratio of remifentanil is approximately tenfold, and the ratio was similar in human studies. Because of these characteristics, more remifentanil can be administered to the mother at times close to delivery than would be considered safe for longer-acting opioids that rely on slower metabolism by the liver. Remifentanil is more effective than long-acting opioids ; however, the improvement in pain relief may be because comparatively larger doses of remifentanil were given.

Although remifentanil may be superior to longer acting systemic opioids, it is inferior to epidural labor analgesia. A meta-analysis of randomized controlled trials comparing remifentanil PCA and epidural analgesia found that parturients with remifentanil PCA had higher pain scores at one hour than those who received epidural analgesia. The incidence of pruritus, nausea, and vomiting were not statistically different between the two groups; however, the confidence intervals were wide. A subsequent meta-analysis confirmed that women were less satisfied when receiving remifentanil PCA compared to women in the epidural group but more satisfied than women receiving other parenteral opioids. The major risk of remifentanil in labor is maternal respiratory depression. Careful surveillance is required to ensure adequate oxygenation and ventilation throughout treatment and for this reason, some practices choose not to offer remifentanil PCA labor analgesia.

Inhaled Analgesia

While volatile anesthetics are no longer used for labor analgesia, nitrous oxide (N ₂ O) is commonly used worldwide. Typically, it is blended with O ₂ in a 50:50 ratio for patient-inhaled self-administration just before and during contractions. A systematic review evaluating N ₂ O for labor analgesia found that epidural analgesia provided more effective pain relief than N ₂ O but most studies were of poor quality. Some studies have demonstrated moderate analgesia in response to inhaled N ₂ O (pain scores of 8 in 10 reduced to 6 in 10), whereas others have shown no difference in visual analogue pain scores. Paradoxically, in this negative study, many women wished to continue using nitrous labor analgesia after the study period. Overall, although patients are less likely to report excellent pain control with N ₂ O, they were as likely to express satisfaction with their anesthesia care. Therefore although it is less efficacious than neuraxial analgesia, it provides an alternative to patients who desire a less invasive analgesic approach as well as those who have a contraindication for neuraxial analgesia. Without the coadministration of opioids, the use of 50% N ₂ O in O ₂ is safe and does not result in hypoxia or unconsciousness.

Neuraxial Analgesia

Neuraxial analgesia is the most reliable and effective method of reducing pain during labor. A large meta-analysis found epidural analgesia offered better pain relief compared with placebo (median pain scale reduction 3.4 in 10, 95% CI, 1.3-5.4) and reduced the need for additional pain medication. Analgesia is given most commonly by epidural, spinal, combined spinal-epidural (CSE), or dural puncture epidural (DPE). These techniques provide superior analgesia to alternative approaches and are safe.

Other Regional Nerve Blocks

Local anesthetic nerve blocks have been used for many years to relieve labor pain, mostly by obstetricians. For a paracervical block, local anesthetic is injected lateral to the cervix at 4 o’clock and 10 o’clock, taking care to avoid vascular structures. It controls pain of the first stage of labor only and is more effective than placebo or intramuscular meperidine. No difference was found in pain relief comparing paracervical block to PCA with intravenous fentanyl. Paracervical block can be complicated by injection of local anesthetic into the presenting fetal head, which can have devastating consequences. More commonly, side effects of transient fetal bradycardia and maternal local anesthetic toxicity have been reported. Therefore for the delivery of a viable fetus, obstetricians in the United States avoid this procedure. The technique of paracervical block is still used in intrauterine fetal demise labor analgesia, dilation and curettage, and dilation and evacuation procedures. It has become safer with more superficial injection ensured by a needle guide and more dilute solutions of local anesthetic.

The pudendal nerve is derived from sacral nerve roots (S2-S4) and can be blocked with local anesthetic using a transvaginal or transperitoneal approach to treat pain during the second stage of labor and for episiotomy repair. Although a pudendal nerve block provides some relief during second stage, it is not as effective as a subarachnoid block with fentanyl and bupivacaine. A pudendal block can impede the urge to push during the second stage of labor. Other complications include a high rate of block failure; systemic local anesthetic toxicity; ischiorectal or vaginal hematoma; and, rarely, fetal injection of local anesthetic.

Maternal Risks and Considerations

Cesarean delivery rates in the United States increased by 50% between 1998 and 2016, rising from 22% to 32% of all births. Common indications for cesarean delivery include fetal malpresentation, nonreassuring fetal status, labor dystocia, and prior cesarean delivery. Although maternal mortality substantially decreased during the first half of the twentieth century, the maternal mortality ratio has not declined in over 25 years and appears to have recently been increasing in the United States. Of interest, the incidence of maternal mortality is 10 times higher in women who underwent cesarean delivery versus vaginal delivery, based on a retrospective study of 1.5 million deliveries between 2000 and 2006.

Pregnant women who undergo general anesthesia for cesarean delivery are at increased risk of pulmonary aspiration of gastric contents, failed intubation of the trachea, or inadequate postoperative ventilation compared with those under neuraxial blockade, particularly in emergent situations. However, it appears that the risks associated with general anesthesia have decreased significantly over time to the point where it is difficult to say that avoiding general anesthesia prevents maternal mortality. Data in the United States from 1979 to 1990 noted a risk ratio of 16.7 for mortality with general anesthesia for cesarean delivery compared with neuraxial blockade, whereas between 1997 to 2002, the risk ratio was not significantly greater with general versus neuraxial anesthesia (RR 1.7; CI, 0.6-4.6, P = .2). This reduction in the risk ratio may be because of advanced airway techniques (e.g., supraglottic airways and video laryngoscopy) and safer airway practices (e.g., difficult airway algorithms).

Use of neuraxial anesthesia for cesarean delivery minimizes exposure of the neonate to maternal anesthetic medications, avoids airway manipulation, improves postoperative pain, and allows the mother to see the child almost immediately after birth. All pregnant women should undergo a preoperative evaluation, regardless of planned delivery mode or type of anesthetic technique, with appropriate risk and benefit counseling. The current status of the fetus and obstetric management plan also should be taken into consideration when formulating the anesthetic plan. In addition, appropriate equipment and medications should always remain readily available to safely provide general anesthesia for an emergent or unanticipated situation.

Although the rates of significant maternal aspiration of gastric contents with induction of general anesthesia are difficult to determine, the mortality from such an event is estimated at 5% to 15% based on retrospective data. ASA guidelines recommend aspiration prophylaxis with the administration of nonparticulate antacids, H ₂ receptor antagonists, and/or metoclopramide before obstetric surgical procedures. The decision to use general anesthesia or neuraxial block for cesarean delivery is determined by a variety of factors that include the fetal condition and urgency of delivery, maternal comorbidities, presence of a previously placed epidural for labor analgesia, surgical considerations, and maternal wishes. At present, most cesarean deliveries in developed countries are performed with neuraxial techniques.

Spinal Anesthesia

If an epidural catheter is not already placed, spinal anesthesia is typically used for nonemergent cesarean deliveries. Compared with an epidural, a single injection spinal is often faster and technically easier to perform, allows adequate operating conditions in a shorter time, provides a denser block, is more cost effective, and is less likely to fail (failure rate <1%). Spinal anesthesia is most often administered through a small (24 gauge or smaller), pencil-point needle. On occasion, a continuous spinal catheter may be used for anesthesia for cesarean delivery. As previously discussed, a spinal catheter may be placed in the case of an inadvertent dural puncture but can be placed intentionally for cesarean delivery in high-risk obstetric patients.

The chance of significant maternal hypotension is greater with spinal anesthesia than with epidural anesthesia. Left uterine displacement with appropriate administration of fluids and use of vasopressor medications can minimize the associated hypotension. Intravenous administration of crystalloid or colloid can reduce the degree of hypotension after spinal anesthesia for cesarean delivery. A Cochrane review assessed 11 trials with 698 women comparing colloid versus crystalloid and found significantly fewer women became hypotensive after colloids (RR 0.68; 95% CI, 0.52-0.89). However, concern exists regarding the safety of synthetic colloids as part of intraoperative and critical care resuscitation. Of note, fluid co-loading is thought to have limited efficacy in consistently preventing postspinal hypotension and is typically utilized in combination with a vasopressor.

Historically, ephedrine was considered the vasopressor of choice to manage hypotension caused by neuraxial anesthesia in pregnancy; however, prophylactic or therapeutic phenylephrine in boluses or as an infusion is not only effective in reducing hypotension but also has less transfer to the fetus and results in less fetal acidosis than ephedrine. Phenylephrine is now considered the vasopressor of choice for the treatment of spinal hypotension, and there is accumulating evidence that administration by prophylactic infusion is most effective in preventing hypotension. A systematic review found prophylactic phenylephrine infusions compared to placebo significantly reduced the risk of hypotension (RR 0.36; 95% CI, 0.18-0.73) and nausea and vomiting (RR 0.39; 95% CI, 0.17-0.91). Phenylephrine is an α-adrenergic receptor agonist and is often associated with reflexive slowing of maternal heart rate and a decrease in cardiac output. There is increasing interest in norepinephrine as an alternative vasopressor for treating spinal hypotension. Compared to phenylephrine, norepinephrine had similar efficacy for maintaining arterial blood pressure during spinal anesthesia for cesarean delivery and was associated with a greater heart rate and cardiac output. Further work needs to be done to assess the safety and efficacy of norepinephrine as the vasopressor of choice to prevent and treat postspinal hypotension.

Although various local anesthetics can be used for spinal blockade, hyperbaric bupivacaine 10 to 12 mg is frequently used to achieve an adequate (T4) level block. Neither patient height nor weight affect block extension, although dosing may require adjustment at extremes of the height spectrum. Lipid soluble opioids (such as fentanyl or sufentanil) may be added to enhance neuraxial blockade by reducing local anesthetic dose and decreasing stimulation from surgical traction of the viscera. Epinephrine (0.1-0.2 mg), may be added to improve the quality and duration of the block. Clonidine can also prolong the duration of the block and improve intraoperative analgesia but can increase sedation and is considered off-label for this use. Preservative-free morphine 0.1 to 0.2 mg or hydromorphone 0.75 mg is frequently administered with the spinal to reduce postoperative pain for 18 to 24 hours after the anesthetic has dissipated.

Epidural Anesthesia

If an epidural catheter has already been inserted for labor analgesia, it provides an excellent method to provide surgical anesthesia for cesarean delivery. A catheter-based technique allows for the ability to titrate the local anesthetic to the proper block height and provide additional local anesthetic administration during the case. For patients who do not already have a catheter in place, this technique may be chosen if the procedure is anticipated to take additional time, or if maternal comorbidities would favor a more gradual, controlled onset of epidural anesthesia. Achieving surgical block conditions takes longer with an epidural than spinal technique but can be rapid enough for use in many urgent situations if already in place and used for maternal analgesia.

Although use of a rapid-onset local anesthetic such as 3% 2-chloroprocaine to attain a T4 level in a newly placed catheter may take 10 minutes, extension of a T10 analgesic level to a surgical anesthetic T4 level can be obtained in approximately 5 minutes using 3% 2-chloroprocaine or alkalinized 2% lidocaine. Bupivacaine 0.5% may be used if rapid onset is not needed, but is often avoided because of the increased risk of local anesthetic systemic toxicity. Typical epidural local anesthetic volumes required for cesarean delivery range between 10 and 20 mL, depending on whether the epidural is already in use. The administration of epidural local anesthetic should occur in divided doses to ensure that the catheter has not migrated into the intravascular or intrathecal space. Block quality can be improved with addition of epinephrine 1:200,000, fentanyl 50 to 100 μg, or sufentanil 10 to 20 μg. Epidural clonidine 50 to 100 μg can be useful in patients with preexisting chronic pain or severe hypertension if the benefit is judged to outweigh the risk for hypotension, bradycardia, and sedation. Epidural morphine 2 to 5 mg is frequently administered to improve postoperative pain.

Combined Spinal-Epidural Anesthesia

Use of a CSE for cesarean delivery may be optimal in some situations as it combines the benefits of the spinal and epidural techniques. This technique allows for the rapid onset of a dense reliable block while allowing the block time or height to be extended with use of the epidural catheter. A low-dose sequential CSE, a technique where a small intrathecal dose of local anesthetic is given followed by administration of epidural medications, can be used in patients with cardiac disease or patients with short stature. Possible disadvantages of a CSE for surgical anesthesia include the presence of an untested catheter and the possibility of a misplaced or nonfunctioning epidural. More details regarding the CSE neuraxial block are discussed in the previous section on labor analgesia.

General Anesthesia

Although neuraxial anesthesia is typically preferred, in certain emergent situations (e.g., fetal bradycardia, maternal hemorrhage or coagulopathy, uterine rupture, maternal trauma) general anesthesia may be needed for cesarean delivery because of its rapid onset. In addition, it allows for a controlled airway, controlled ventilation, and in some scenarios such as massive hemorrhage, improved hemodynamic control and perhaps decreased maternal psychological stress in comparison to neuraxial anesthesia.

Appropriate equipment preparation, knowledge of patient comorbidities, airway examination, and familiarity with the difficult airway algorithm are necessary preparation for delivering a safe general anesthetic. The ASA difficult airway algorithm has been modified slightly by some authors for use in the setting of cesarean delivery, and a previously published example is provided in Fig. 62.4 . Clear, concise communication among all members of the perioperative team is especially critical in urgent or emergent situations to maximize patient safety and minimize procedural complications. Open lines of communication are essential around the time of induction of anesthesia, airway management, and surgical incision.

Algorithm for management of unanticipated difficult airway in obstetric patients.

BMV, Bag-mask ventilation; BP, blood pressure; CS, cesarean section; ETCO ₂ , end-tidal carbon dioxide; HR, heart rate; LMA, laryngeal mask airway; SpO ₂ , oxygen saturation.

Redrawn from Balki M, Cooke M, Dunington S, et al. Unanticipated difficult airway in obstetric patients: development of a new algorithm for formative assessment in high-fidelity simulation. Anesthesiology . 2012;117:883–897, with permission.

A rapid-sequence induction commences with preoxygenation, followed by the application of cricoid pressure and the administration of an intravenous induction drug (typically propofol) and a neuromuscular-blocking drug (typically succinylcholine or rocuronium). If endotracheal intubation fails, consider placement of a supraglottic airway device, such as a laryngeal mask airway (LMA), or ventilation with mask and cricoid pressure. The LMA has a high success rate for placement in obstetric patients; however, it does not protect against aspiration of gastric contents and should be used primarily for rescue of a failed intubation. Of note, the LMA has been used without observed aspiration or presence of hypoxia in a prospective study of over 1000 lean, low-risk patients undergoing elective cesarean delivery.

Complications as a result of general anesthesia for cesarean delivery can be more common than general anesthesia for other procedures. For example, although the overall risk of intraoperative awareness is estimated to be 1:19,000 general anesthetics, the awareness risk for cesarean delivery is estimated at 1:670 (1:380-1:1300). Anesthesia-related mortality from airway difficulties may occur during emergence and in the recovery period, as well as during induction. Improper monitoring, provider inexperience, emergent situations, and patient obesity all increase patient risk. Emergent cesarean delivery requiring general anesthesia is an uncommon but predictable emergency that can be practiced by teams using simulation to improve performance.

Postdural Puncture Headache

The earliest experiments with spinal cocaine resulted in severe PDPH. Leakage of spinal fluid is thought to result in vascular hyperemia, migraine physiology, and traction on pain-sensitive fibers. The headache associated with PDPH is postural in that it is worsened by standing and relieved by lying down. The incidence, severity, and duration of PDPH are related to the size of the needle and the shape of the tip. Spinal needles used for CSE technique range from 25 to 29 gauge and result in an incidence of PDPH of less than 1%. Epidural catheters are most commonly placed through a 17- or 18-gauge blunt-tipped needle. The incidence of unintentional dural puncture during labor epidural placement is 1% to 1.5%. The incidence of headache after an unintentional dural puncture with an epidural needle is reported at 30% to 80%.

In the setting of an unintended dural puncture with an epidural needle, an intrathecal catheter may be threaded, or the epidural needle may be removed and replaced at a different interspace. If an intrathecal catheter is placed, unintentional injection of an epidural anesthetic dose must be carefully avoided. Placement of the intrathecal catheter can provide labor analgesia and alleviates the need for multiple repeat epidural attempts with the potential of a second accidental dural puncture.

When diagnosing a PDPH, it is important to consider other causes of headache in the postpartum period. Assessing the patient for fever and nuchal rigidity is important because postdural-puncture meningitis can initially present with a headache. Early treatment of meningitis is important to prevent morbidity and mortality. Likewise, assessing the patient for hypertension is important to detect postpartum preeclampsia, which can present with a headache and requires rapid treatment to prevent maternal stroke. A thorough neurologic exam should also be performed because cerebral venous thrombosis, cranial subdural hematoma, and ischemic or hemorrhagic stroke can present as a postpartum headache. Benign headaches from tension, dehydration, sleep deprivation, caffeine withdrawal, or migraine should be considered prior to treating a PDPH.

PDPH is often initially treated conservatively. Given the similarity of symptoms to those of migraine headache, PDPHs have been treated with drugs that are useful in migraine with variable success. Caffeine can be minimally effective to treat the pain of a PDPH in the short-term likely because of its vasoconstrictive effects.

If the symptoms are severe enough to limit a patient’s activity then an epidural blood patch (EBP) should be considered. If there is evidence of cranial nerve involvement such as diplopia, then an EBP should be performed immediately. Note that the symptom of muffled hearing is common with PDPH and thought not to be a result of cranial nerve involvement, but instead, a decrease in middle ear pressure because the middle ear fluid is connected to the cranial CSF via the cochlear aqueduct. In one retrospective review of an EBP database, all parturients with PDPH experience relief after the EBP but 16.8% required two and 1.5% required three EBPs. Controversy exists over the optimum volume of blood to be injected during an EBP but studies support an attempt to administer 20 mL of autologous blood during the procedure.

Epidural Hematoma

The epidural space is highly vascular, and vessels can be punctured during neuraxial needle placement or catheter threading. However, with normal platelets and coagulation factors, epidural hematoma is extremely uncommon. The Serious Complication Repository (SCORE) Project reported the incidence of epidural hematoma to be 1 in 251,463 (95% CI, 1:46,090-1:10,142,861) and another large study found no cases of hematoma out of 79,837 obstetric epidural placements with an estimated upper limit of 1 in 4.6 × 10 ^−5. Although rare, back pain and persistent motor blockade are potential signs of epidural hematoma and should be thoroughly evaluated. Patients should be followed postpartum until complete resolution of the epidural blockade. Patients are at increased risk for developing a hematoma in the setting of coagulopathy and anticoagulation use. Following the guidelines of the American Society of Regional Anesthesia (ASRA) with consideration for the Society for Obstetric Anesthesia and Perinatology (SOAP) Consensus Statement ( Fig. 62.5 ) regarding anticoagulation therapy and neuraxial anesthesia in the obstetric patient is recommended. If an epidural hematoma is suspected, imaging should occur immediately with the goal of prompt epidural hematoma evacuation to avoid permanent neurologic injury.

(A) Decision aid for urgent or emergent neuraxial procedures in the obstetric patient receiving unfractionated heparin. (B) Decision aid for urgent or emergent neuraxial procedures in the obstetric patient receiving LMWH. aPTT, activated partial thromboplastin time; GA, general anesthesia; LMWH, low-molecular-weight heparin; SEH, spinal epidural hematoma; SQ, subcutaneous; UFH, unfractionated heparin.

Reproduced with permission from Leffert L, Butwick A, Carvalho B, et al. The Society for Obstetric Anesthesia and Perinatology Consensus Statement on the Anesthetic Management of Pregnant and Postpartum Women Receiving Thromboprophylaxis or Higher Dose Anticoagulants. Anesth Analg. 2018;126:928–944.

Neurologic Injury

Direct spinal cord damage from an epidural or spinal needle placed for labor is exceedingly rare because obstetric neuraxial anesthesia procedures are typically performed below the level of the conus medullaris. It has, however, been described with spinal anesthesia attempts performed at unintentionally high levels resulting in spinal cord syrinx formation with injection. In a series of seven such cases, all of the patients who experienced this complication described pain on injection. Therefore if a patient complains of pain on injection of neuraxial medication, the proceduralist should stop injecting immediately .

Overall, damage attributed to direct neurologic injury has been estimated at 0.6 in 100,000 with epidural and 3 in 100,000 with spinal analgesia. Data from the SCORE project show an incidence of serious neurologic injury in the postpartum period to be 1 in 11,389 (95% CI, 1:7828-1:17,281) but only 1:35,923 (95% CI, 1:17,805-1:91,244) are thought to be related to anesthesia.

Evaluating postpartum lumbosacral neuropathies is a common part of an obstetric anesthesia practice. A large prospective study showed that 0.92% of women experienced postpartum lumbosacral or lower extremity nerve injury, the incidence of which was associated with nulliparity and prolonged second stage of labor, but not epidural analgesia use. Femoral and lateral femoral cutaneous neuropathies were the most common, likely from women in extreme hip flexion in the semi-Fowler position. Encouraging women to straighten their legs between pushes can restore blood flow to the nerves of the lumbar plexus. Most of these peripheral neuropathies heal with time but outpatient follow up with a neurologist to follow the course and rule out other causes should be considered.

Local Anesthetic Systemic Toxicity

The accidental injection of intravascular local anesthetic could result in the development of local anesthetic systemic toxicity (LAST). The development of LAST is rare in obstetrics secondary to the dilute local anesthetic solutions utilized for epidural analgesia and the use of lidocaine and 2-chloroprocaine for operative delivery. The SCORE project reported one maternal cardiac arrest from LAST after a TAP block. LAST as a result of TAP blocks in obstetrics has been reported multiple times and the use of no greater than 0.25% concentration of bupivacaine, adding 1:200,000 epinephrine, administering no greater than a 20 mL volume on each side, placement under ultrasound to be sure the injection is not intraperitoneal, and careful, slow, injection with intermittent aspiration is recommended. The treatment of LAST includes the administration of lipid emulsion in addition to basic and advanced cardiac life support.

Total Spinal Block

A total spinal block is a rare and life-threatening complication that occurs after excessive cephalic spread of local anesthetic in the CSF resulting in severe respiratory and cardiac compromise. It can occur after a single spinal injection or as a result of inadvertent intrathecal spread of epidural medication. Risk factors associated with high neuraxial blockade include obesity, spinal technique after failed epidural anesthesia, short stature, epidural after an accidental dural puncture, and spinal deformity.

Other Complications

When strict aseptic technique is used, infection is uncommon with spinal and epidural anesthesia. Nonetheless, postdural puncture meningitis and epidural abscess has been described and the ASA and ASRA recommend the following aseptic techniques during placement of neuraxial needles and catheter: removal of jewelry, hand washing, wearing of caps, wearing of masks covering both mouth and nose, use of sterile gloves, use of an antiseptic solution (e.g., chlorhexidine with alcohol), and sterile draping of the patient.

Postpartum back pain is common after childbirth regardless of whether neuraxial analgesia has been used. However, no evidence indicates that back pain is more common when neuraxial analgesia is used for labor. Several trials have identified an association between increased maternal temperature and epidural use as a secondary outcome. Attribution of causality is difficult in this setting and the etiology is not well understood but likely involves noninfectious inflammation.

Hypertensive Disorders

Hypertensive disorders of pregnancy are among the most common causes of maternal morbidity and are associated with more frequent rates of maternal and fetal mortality. Hypertensive disorders of pregnancy complicate 5% to 10% of pregnancies worldwide and preeclampsia is diagnosed in 3% of pregnancies. The World Health Organization (WHO) has identified hypertension as the second most common cause of maternal death, accounting for 14% of mortality associated with pregnancy. Chronic hypertension may precede pregnancy and may or may not be complicated by superimposed preeclampsia.

Although some variations exist in hypertensive definitions worldwide, the following are used in the United States by the ACOG based on a working group in 2013. Gestational hypertension is defined as the onset of hypertension (systolic blood pressure [SBP] >140 mm Hg or diastolic blood pressure [DBP] >90 mm Hg) after 20 weeks’ gestation in a previously normotensive parturient without proteinuria. Preeclampsia is defined as hypertension (SBP >140 mm Hg or DBP >90 mm Hg) after 20 weeks’ gestation associated with proteinuria. Preeclampsia is diagnosed when urine protein excretion is greater than 300 mg in a 24 hour period or, alternatively, there is a protein/creatinine ratio of at least 0.3. In 2013, massive proteinuria (>5 g in 24 hours) and fetal growth restriction were eliminated as considerations of severe preeclampsia. In addition, the term mild preeclampsia is no longer used. Now only preeclampsia or preeclampsia with severe features are defined. Severe features of preeclampsia include an SBP of 160 mm Hg or greater or DBP of 110 mm Hg or greater on two separate occasions at least 4 hours apart while on bed rest; thrombocytopenia (platelet count less than 100,000/μL); impaired liver function with twice normal concentrations of liver enzymes; right upper quadrant pain; progressive renal insufficiency with serum creatinine greater than 1.1 mg/dL or a doubling of serum creatinine without other known renal disease; pulmonary edema; and new onset cerebral or visual abnormalities. In the absence of proteinuria, preeclampsia can be diagnosed with new onset hypertension (as previously defined) and the presence of a severe feature.

The combination of hemolysis, elevated liver enzyme, and low platelet count (HELLP) is considered preeclampsia with HELLP syndrome. Eclampsia is preeclampsia complicated by seizure activity. The incidence of preeclampsia has increased likely as a result of increases in maternal age and obesity, whereas the risk for eclampsia has decreased because of improved prenatal care and the use of prophylactic intravenous magnesium. The cause of preeclampsia is not known, but it is generally thought that placental insufficiency through a cascade of antiangiogenic factors (e.g., s-Flt) eventually leads to maternal generalized endothelial dysfunction. The cause of the placental insufficiency is likely variable and could include maternal or paternal genetics and/or environmental factors. Patients with preeclampsia have an elevated risk for cerebral hemorrhage, pulmonary edema, and coagulopathy. Current guidelines recommend treating SBP more than 160 mm Hg for prevention of intracerebral hemorrhage. Initial treatment typically includes intravenous labetalol, hydralazine, and/or oral nifedipine. Other considerations are increased airway edema with associated difficulty of intubation and increased rates of postpartum atony related to the use of magnesium sulfate. Methylergonovine (Methergine) should be used cautiously in patients with preeclampsia because it may lead to hypertensive crisis. Women with preeclampsia can be sensitive to both endogenous and exogenous catecholamines. Therefore careful administration of adrenergic agents is recommended.

Women with a diagnosis of preeclampsia should have their platelet count checked before initiation of regional anesthesia or removal of an epidural catheter. Coagulopathy is a contraindication to regional anesthesia. Although the risk for spinal hematoma is lower in pregnant women than in the elderly, one study found 68% of patients who had spinal hematomas after neuraxial blockade had preexisting coagulopathy. Lastly, despite the concerns for hypotension, spinal anesthesia is safe to administer in preeclamptic patients.

Coagulopathies

Thrombocytopenia complicates 10% of pregnancies as a result of several etiologic factors. Thrombocytopenia may be preexisting or can develop as a result of the pregnancy. As discussed earlier, thrombocytopenia that develops after 20 weeks’ gestation may be a sign of preeclampsia with HELLP syndrome. However, most thrombocytopenia that develops in pregnancy is benign, gestational thrombocytopenia. The platelet count is expected to decrease by approximately 10% in a normal pregnancy. Autoimmune thrombocytopenia, antiphospholipid syndrome, and liver disease are less common. Glucocorticoids and IV immunoglobulin can be used to elevate platelet counts in certain types of severe disease but require several days to be effective. No platelet count is universally accepted as safe for epidural placement. Most anesthesiologists agree that placement of an epidural in the setting of a platelet count greater than 100,000/mm ³ is safe and recent literature suggests lower thresholds may be safe. A retrospective observational study of cases from 14 diverse institutions combined with a systematic review evaluated 1524 parturients with a platelet count of less than 100,000/mm ³ and found the upper bound of the 95% CI for the risk of epidural hematoma for a platelet count of 0 to 49,000/mm ³ is 11%, 50,000 to 69,000/mm ³ is 3%, and 70,000 to 100,000/mm ³ is 0.2%. Of note, no cases of epidural hematomas requiring surgical decompression were identified in this cohort.

Women with von Willebrand disease are at increased risk for bleeding intrapartum and postpartum. Prophylactic treatment is recommended for women with von Willebrand factor (vWF) less than 50 international units/dL. Because of the multiple types and subtypes of von Willebrand disease that have different responses to therapy, it is imperative that hematologic studies be part of the management to help guide the most appropriate therapy. In type I von Willebrand disease, a partial reduction in the quantity of vWF is seen. Women with type I von Willebrand disease usually do not need prophylactic treatment because their vWF typically rises in the third trimester of pregnancy to greater than 50 international units/dL. Desmopressin can be used in women with type 1 von Willebrand disease to release vWF. Type 2 von Willebrand disease is characterized by a defect in the function of vWF, and facilitation of release is not helpful. Parturients with type 3 von Willebrand disease almost always require replacement of vWF before delivery, because they have almost no intrinsic vWF. Although concern should be elevated for epidural hematoma, women with normal factor levels can have regional anesthesia in the setting of a normal platelet count.

Deep venous thrombosis (DVT) and pulmonary embolism are more common in pregnancy as a result of hormonal changes. Significant risk factors are factor V Leiden, prothrombin G20210A, protein S, protein C and antithrombin deficiency, and antiphospholipid antibodies. Patients diagnosed with DVT or pulmonary embolism in pregnancy will need prolonged anticoagulation and will require delivery planning with a brief anticoagulation hiatus for regional anesthesia and delivery. Factor V Leiden is an abnormal variant of factor V that acts as a cofactor that allows activation of thrombin by factor Xa. The factor V Leiden variant cannot be easily degraded by activated protein C and thus leads to hypercoagulability. Patients with a factor V Leiden abnormality may be maintained on anticoagulants for prophylaxis or treatment of DVT and appropriate cessation of anticoagulants must occur prior to placement of neuraxial blockade.

Obesity

Maternal obesity (prepregnancy body mass index ≥30 kg/m ² ) and metabolic syndrome cause gestational diabetes, newborn hyperglycemia, larger babies, prolonged labor, and cesarean delivery. Cesarean delivery for obese women holds greater risk of mortality. Sleep apnea is more common in obese parturients and is a risk factor for hypoventilation after treatment with systemic opioids and difficult tracheal intubation when general anesthesia is required. Morbidly obese parturients are at increased risk of longer first stage of labor and operative delivery. Epidural anesthesia placement in obese parturients is more difficult and takes longer and they are more likely to require a repeat procedure due to inadequate pain control or failure to achieve bilateral sensory blockade. An early anesthesia consultation is advised for morbidly obese women regardless of planned delivery mode.

Cardiac Disease

The leading cause of maternal mortality in the United States is cardiac disease. In planning an anesthetic for a delivery in a woman with congenital or acquired heart disease, the anesthesia provider must take into consideration the patient’s cardiac lesion or disease state; the normal physiologic changes of pregnancy, labor, and delivery; and the hemodynamic changes of the anesthetic itself. During pregnancy, labor, and delivery, regurgitant valvular lesions are generally tolerated better than stenotic valvular lesions. Pulmonary hypertension, severe left ventricular outflow tract obstruction, moderate to severe mitral stenosis, and cyanotic congenital heart disease all carry significant risk for maternal morbidity and mortality. Multiple risk factors have been identified as increasing a woman’s likelihood of major morbidity and mortality including prior stroke, low ejection fraction, aortopathy, and heart failure symptoms. The American Heart Association, the American College of Cardiology, and the European Society of Cardiology have classified certain conditions or cardiac lesions as high maternal or fetal risk. In comparison to other risk stratification systems, the modified WHO risk stratification system appears to perform best in predicting maternal morbidity or mortality during pregnancy and delivery. Such risk stratification informs the obstetric and anesthesia team as to what resources a parturient with heart disease may require, and thereby whether they need to transfer to a tertiary care facility for delivery.

Increased monitoring including 5-lead ECG or intraarterial blood pressure monitoring may be necessary for labor, especially in women who have a history of tachyarrhythmia or are hemodynamically tenuous. Epidural labor analgesia is recommended for women with heart disease to decrease catecholamine release and eliminate the increased cardiac output and tachycardia attributable to labor pain. The afterload reduction that occurs with epidural analgesia should be followed closely and may need to be countered with a carefully titrated α-adrenergic agonist to prevent tachycardia and ischemia. When cesarean delivery is indicated, it is important to tailor the anesthetic technique to each individual patient. Regional anesthesia is not necessarily contraindicated and should be considered for most patients.

Anticoagulation

Anticoagulation is often required for women with artificial heart valves, certain types of congenital heart disease, pulmonary hypertension, and cardiomyopathy. It is necessary to carefully time stopping anticoagulation to allow for epidural placement and birth while restarting before a clot can form. For this reason, women are usually maintained on heparin at the end of pregnancy, pending delivery, because of its quick offset. If IV unfractionated heparin was used, it should be stopped 4 to 6 hours before the procedure and normal coagulation status (PTT or activated clotting time [ACT]) documented before neuraxial instrumentation or removal of an epidural catheter. If a parturient is maintained on subcutaneous unfractionated heparin, the time interval required to stop prior to a neuraxial procedure corresponds to the dose and activated partial thromboplastin time (aPTT) that can be measured (see Fig. 62.5 A ). If conversion to intravenous heparin cannot take place before labor, warfarin should be stopped and neuraxial analgesia deferred until the PT is within normal limits and the INR is documented at less than 1.5. Low-molecular-weight heparin (LMWH) has been used frequently for prophylaxis of DVT in pregnancy. Unlike unfractionated heparin, the anticoagulant effects of LMWH cannot be reliably measured, and it is not reversible with protamine. It is recommended to wait a prespecified time interval prior to performing a neuraxial procedure, rather than ordering laboratory testing. The decision to place or remove a neuraxial block in a patient on LMWH should correspond to the individual dose and total daily dose (see Fig. 62.5 B ). Nonsteroidal antiinflammatory drugs do not in themselves increase risk for spinal hematoma, but can increase risk in combination with other anticoagulants. If labor begins before neuraxial analgesia can safely be given, intravenous remifentanil and N ₂ O are alternatives in some settings.

Pulmonary Disease

As previously detailed in the section on pulmonary changes, adaptations in the respiratory system are required to meet the metabolic demands of the mother and fetus with increases in minute ventilation, decreased O ₂ reserve, and increased airway edema most notable.

Asthma is characterized by reversible airway obstruction, airway hyperresponsiveness, and airway inflammation. It is the most common respiratory disease in pregnancy with considerable maternal morbidity. A prospective trial of 1739 pregnant asthmatic patients found those with mild asthma had an exacerbation rate of 12.6% and a hospitalization rate of 2.3%, those with moderate asthma had an exacerbation rate of 25.7% and a hospitalization rate of 6.8%, and those with severe asthma had an exacerbation rate of 51.9% and a hospitalization rate of 26.9%. Bronchodilators and antiinflammatory drugs are generally safe for the fetus and should be used in pregnancy to control asthma. A meta-analysis found that maternal asthma was associated with increased risk of maternal and placental complications including cesarean delivery, gestational diabetes, abruption, and hemorrhage.

Community-acquired pneumonia is the most common nonobstetric infectious complication that results in maternal death. During the 2009 H1N1 influenza epidemic, pregnant women were disproportionately affected. Pregnant women were at increased risk for hospitalization, ICU admission, and death. Preterm delivery is a significant complication of pneumonia even in the setting of antibiotic therapy. Aspiration of gastric contents is also more common during endotracheal intubation in pregnant than in nonpregnant patients because of laxity of the gastroesophageal junction combined with anatomic changes from the expanding uterus. Strict fasting recommendations, rapid-sequence tracheal intubation, and antacids are used before general anesthesia in pregnancy.

Cystic fibrosis is a common autosomal-dominant disorder in women of northern European origin. With improved medical care, women with cystic fibrosis are living to reproductive years and beyond. Pregnancy is uncommon in women with cystic fibrosis (216 pregnancies in 24,000 women), but when it occurs, careful multidisciplinary care is required. The disease is caused by a mutation in the cystic fibrous membrane conductance regulator in epithelial cells, which causes abnormalities in the lungs, pancreas, intestines, and hepatobiliary systems. The major issues in pregnancy are restrictive lung disease and diabetes.

Neurologic Disorders

Multiple sclerosis is a neuroinflammatory disorder that disproportionately affects young women. Pregnancy is associated with a decrease in the incidence of relapse, although the rate of relapse during the first 3 months postpartum is increased in comparison with the year before pregnancy. Multiple sclerosis is a disease of demyelination, and thus theoretic concern exists that local anesthetic toxicity may be enhanced. Cases of worsening symptoms after regional anesthesia have been reported; however, it is hard to impute causality in a relapsing and remitting disease. Nevertheless, the lowest effective concentration possible should be given and vasoconstrictive agents should be avoided. Some recommend epidural instead of intrathecal local anesthetic administration when possible.

Neurofibromatosis is an autosomal-dominant disorder that occurs in 1 in 3000 individuals, with variable manifestations. It is characterized by café-au-lait lesions on the skin, cutaneous neurofibromas, Lisch nodules of the iris, bone abnormalities, and tumors of the spinal cord and cranial nerves. Neurofibromas often grow in pregnancy. Disagreement exists as to whether neuraxial anesthesia is contraindicated in women with neurofibromatosis because of the incidence of vascular spinal tumors. Epidural hematoma has been reported in a woman with neurofibromatosis in the setting of a spinal tumor. The hormonal changes of pregnancy may cause tumor growth, and a knowledge of lesion location and current clinical symptoms is needed to avoid instrumentation of tumors and safely deliver neuraxial anesthesia.