Anesthesia in High-Risk Obstetrics
Cynthia A. Wong
Table Of Contents
Cynthia A. Wong, MD
NONOBSTETRIC SURGERY DURING PREGNANCY|
Anesthesiologists often contribute to the care of obstetric patients at high risk. The anesthetic care of parturients at high risk in the peripartum period has been discussed elsewhere. This chapter reviews anesthetic issues involved in caring for women at high risk with diseases or conditions unrelated to their pregnancy that complicate their obstetric or their obstetric anesthesia care. Often, these women have special anesthetic considerations. Appropriate anesthetic management can assist in the obstetric management of these women. Antepartum consultation between the obstetrician, anesthesiologist, and specialist managing the pregnant woman's chronic condition will help assure the best outcome possible for both the mother and her child(ren).
|NONOBSTETRIC SURGERY DURING PREGNANCY|
Maternal physiologic and anatomic changes that accompany pregnancy, as well as consideration of the developing fetus, influence the conduct of anesthesia for nonobstetric surgery during pregnancy. Physiologic changes associated with pregnancy may affect maternal safety during anesthesia. Fetal oxygenation depends on maternal oxygen carrying capacity, maternal cardiac output, and uteroplacental perfusion. Therefore, any interventions that compromise these factors may lead to fetal asphyxia. It is often difficult to determine whether adverse fetal outcomes are the result of maternal disease, the surgical procedure, or the anesthetic.
Changes in the cardiovascular system during pregnancy need to be considered when caring for the pregnant surgical patient. Pregnant women respond differently than nonpregnant patients to hemorrhage. Compression of the great vessels by the gravid uterus after 20 weeks' gestation may decrease maternal cardiac output and uteroplacental blood flow.1 After 20 weeks' gestation the pregnant patient should not undergo operations or be nursed in the supine position. Because maternal blood volume is increased and hematocrit is decreased, pregnant women can tolerate a higher percentage of blood loss before becoming symptomatic. For example, pregnant ewes acutely lost 30% to 35% of estimated blood volume with no change in maternal blood pressure.2 Thus, the degree of hypovolemia may initially be underestimated.
Changes in respiratory anatomy and physiology must also be considered. Evaluation of maternal oxygen carrying capacity must take into account the increased maternal metabolic rate (approximately 50% at term3) and fetal oxygen requirements. Maternal functional residual capacity (FRC) is decreased by 20%.4 General, high-spinal, and high-epidural anesthesia are associated with a further decrease in FRC.5,6 Therefore, oxygen reserve is compromised in the anesthetized pregnant patient. Even short periods of apnea (e.g., during induction of general anesthesia) may result in rapid and severe hypoxemia, leading to both maternal and fetal compromise.7 After upper abdominal and thoracic procedures, FRC continues to decrease to a nadir at 48 hours postoperatively.8 Therefore, consideration should be given to monitoring maternal oxygenation or providing supplemental oxygen during this time. Maternal hyperoxia is not an issue, because even with maximal maternal pO2 levels, fetal pO2 never increases to more than 60 mm Hg.9
Carbon dioxide readily crosses the placenta; therefore, maternal pCO2 levels reflect fetal pCO2 levels. Maternal hypoventilation results in increased fetal pCO2 and fetal respiratory acidosis. Maternal hyperventilation shifts the oxyhemoglobin dissociation curve to the left, thus increasing the affinity of hemoglobin for oxygen and decreasing the transfer of oxygen across the placenta.10 In addition, maternal positive-pressure mechanical hyperventilation can lead to decreased venous return and decreased maternal cardiac output.11
Sensitivity to central nervous system depressants (including inhaled12,13 and intravenous agents14 and local anesthetics15) is increased during pregnancy. Drug volume of distribution may be altered during pregnancy.16 In addition, decreases in maternal protein binding may cause an increase in the free drug concentrations of highly protein-bound drugs.17 Therefore, pregnant patients are at increased risk for an anesthetic overdose.
Both anatomic-induced and hormonally-induced changes in the gastrointestinal system cause pregnant women to be at increased risk for gastroesophageal reflux and pulmonary aspiration.18 This is an issue during periods of maternal obtundation (e.g., general anesthesia, conscious sedation, or central nervous system pathology).
Virtually all anesthetic agents administered to the mother cross the placenta to the fetus. This is of particular concern during the period of organogenesis, although the brain continues to develop well after the first trimester. The teratogenicity of most anesthetic agents has not been well studied in humans, because both birth defects and exposure to anesthetic agents during pregnancy are unusual events. Several retrospectives reviews suggest there is no increased risk for congenital anomalies in the children of women who undergo surgery while pregnant, but there is a risk of abortion, growth restriction, and an increased incidence of very-low and low-birth-weight infants.19,20 A review from Swedish health care registries of more than 2200 women who underwent surgery during the first trimester found a possible increase in neural tube defects in their children.21 A case-controlled study in Atlanta found an increased risk of hydrocephalous if surgery was performed in the first trimester.22 However, the studies did not control for indication for surgery, type of surgery, type of anesthesia, or other exposures. None of the anesthetic agents currently used is a known teratogen in humans. However, fetal hypoxia secondary to either anesthetic or surgical procedures is a known risk to the fetus. Therefore, most anesthesiologists recommend that elective surgical procedures be postponed until after the first trimester. Although fetal exposure to drugs is more limited during spinal compared with general anesthesia, there is no evidence that one type of anesthesia is safer than another.
Laparoscopy During Pregnancy
The number of laparoscopic procedures performed during pregnancy is increasing.23 Studies in pregnant ewes have found that maternal blood pressure and cardiac output generally decrease during carbon dioxide pneumoperitoneum, but that fetal cardiac output is maintained. However, animal studies conflict as to the relationship between maternal end-tidal CO2 and pCO2 (and hence, fetal pCO2) during CO2 pneumoperitoneum. Despite controlled ventilation to maintain end-tidal CO2 at a constant level, two studies found maternal and fetal hypercarbia, that is, a large end-tidal to arterial CO2 gradient,24,25 while one-third found no change in maternal and fetal pCO2.26 A small study in pregnant women undergoing laparoscopic cholecystectomy found no increase in the end-tidal to arterial pCO2 gradient during CO2 insufflation.27 Prolonged exposure of pregnant laboratory animals to carbon dioxide results in congenital anomalies in their offspring.28 Although there is no evidence in humans that laparoscopy is associated with an increased risk for congenital anomalies, the Society of American Gastrointestinal Endoscopic Surgeons' Guideline for Laparoscopic Surgery During Pregnancy states that “maternal end-tidal CO2 and arterial blood gases should be monitored.”29
Advanced trauma life support protocols should be used for pregnant trauma patients.30 Women of childbearing age should be evaluated for pregnancy early in trauma management.31 Operative anesthetic considerations are discussed elsewhere in this chapter. In an emergency, O-negative, cytomegalovirus (CMV)-negative red blood cells should be used, and Rh-immunotherapy should be administered if the possibility of uterine injury exists.31 The decision to perform a perimortem cesarean delivery should be made within 5 minutes of maternal to arrest to improve the chances of fetal survival.32
Anesthesia for Fetal Surgery
Fetal surgery may be indicated for selected fetal anomalies and is currently an area of investigation in selected centers. Anesthetic considerations are similar to those for women undergoing nonobstetric surgery during pregnancy. Additional considerations include the need for fetal anesthesia, fetal immobility, and profound uterine relaxation, while maintaining fetal cardiovascular and acid–base status.33,34
The human fetus responds to environmental stimuli. Noxious stimuli result in an autonomic response and an increase in stress hormone levels.35,36 In premature infants undergoing surgery, this stress response is ameliorated by adequate anesthesia.37 Although there is no evidence to support a difference in outcome, it seems intellectually logical and humane to provide fetal anesthesia during fetal surgery.
Volatile anesthetic agents would seem to be the ideal agents to anesthetize both the mother and the fetus for fetal surgery. They provide both fetal anesthesia and immobility and uterine relaxation. However, animal studies of the effects of volatile anesthetic agents on uteroplacental blood flow, fetal blood pressure, heart rate, oxygen status, and acid–base status are conflicting.38,39,40 The cardiovascular effects of volatile anesthetic agents are dose-dependent in both the mother and the fetus,38 and maintenance of maternal cardiovascular stability, as well as other factors affecting oxygen delivery to the placenta, are important.
Other fetal anesthetic techniques have been described, including the direct fetal injection (umbilical vein or intramuscular) of fentanyl and pancuronium.33 Volatile anesthetic-induced uterine relaxation can be supplemented with maternal intravenous administration of nitroglycerine, magnesium or β-mimetics, or rectal indomethacin.33,34 These drugs may interact with anesthetic agents to produce adverse side effects.34 The results of animal studies suggest that controlling maternal postoperative pain decreases myometrial irritability after fetal surgery.41,42 Continuous epidural analgesia with low-concentration bupivacaine and fentanyl is used by several centers performing fetal surgery.33,34
Both the mother and fetus require monitoring during fetal surgery. Standard monitors are used for the mother. Invasive arterial and central venous pressure monitoring, and a Foley catheter, may be indicated if large fluid shifts are anticipated, especially if β-mimetic tocolysis is used, because this increases the risk of maternal pulmonary edema.34 Ultrasonography is used to determine fetal and placental orientation, and to monitor fetal heart rate (FHR) and contractility.33 In the pregnant ewe model, fetal pulse oximetry had a high sensitivity and negative predictive value for monitoring fetal well being in utero.43
Maternal Systemic Disease
Many of the anatomic and physiologic changes of pregnancy are accentuated by obesity. Chest wall compliance is decreased. Oxygen consumption and carbon dioxide production are increased further. Lung volumes are further decreased in obese parturients.44 Oxygenation worsens in the supine or Trendelenburg position,45 often contributing to the difficulty of positioning these patients for surgical procedures. Cephalad retraction of a large panniculus may contribute to maternal hypotension and fetal compromise.46
Similarly, cardiac output and blood volume are increased in obesity and obese patients are at increased risk for hypertension.47,48 It is not clear whether the gastrointestinal changes associated with pregnancy are exacerbated by obesity, although it seems likely that the obese pregnant patient is at increased risk for pulmonary aspiration.45 Obesity influences the distribution, elimination, and response to anesthetic drugs. There is a higher incidence of difficult airway,47 and this, combined with an increased metabolic rate and decreased FRC, makes airway catastrophes more likely. Several studies have found that obesity increases the risk for anesthesia-related maternal mortality during cesarean section.49,50
Obesity does not affect the severity of labor pain.51 However, it is technically more difficult to establish epidural analgesia in the obese parturient and several attempts at catheter placement may be necessary to obtain effective analgesia.47 Epidural analgesia does not affect the outcome of labor in obese parturients weighing more than 300 pounds.47 However, the risk of cesarean delivery is greater in obese parturients.47,52 Therefore, early labor epidural analgesia is recommended.
Even more so than in nonobese women, neuraxial anesthesia is preferable to general anesthesia for cesarean delivery. Spinal anesthesia can be used successfully in morbidly obese women; however, operative times tend to be prolonged,48 and if the duration of surgery extends beyond the duration of spinal anesthesia, the anesthesiologist will be forced to induce general anesthesia intraoperatively. Obese patients are at increased risk for postoperative pulmonary morbidity.53,54 A vertical skin incision increases the risk of postoperative hypoxemia.55 Finally, in the postoperative period, opioid-induced respiratory depression can have more profound effects in the obese patient who has less oxygen reserve.
Asthma is becoming more prevalent in the general and pregnant population.56 During labor and delivery, high-dose opioid analgesia may cause respiratory depression and decompensation in a wheezing patient. Neuraxial analgesia prevents pain-induced hyperpnea without respiratory depression and also ameliorates the maternal stress response.57 This may be important for women who have exercise or stress-induced asthma.58 An additional benefit of epidural analgesia is that epidural anesthesia can be rapidly induced for an emergency cesarean delivery, thus avoiding the need for general anesthesia and intubation. Airway instrumentation is the greatest risk factor for bronchospasm in the perioperative period.59 In nonpregnant asthmatic patients, volatile anesthetics are the agents of choice because they produce dose-dependent bronchodilation60 and suppress airway reflexes.61 However, high concentrations of volatile anesthetic agents also relax uterine, as well as bronchial, smooth muscle62 and may increase the risk of postpartum hemorrhage. In the nonasthmatic pregnant patient, extubation of the trachea when the patient is wide-awake minimizes the risk of pulmonary aspiration. However, the presence of an endotracheal tube in an asthmatic patient who is awake may stimulate bronchospasm. Inhaled beta-adrenergic agents administered before anesthesia emergence may decrease this risk. However, the wheezing patient under general anesthesia should probably remain sedated and intubated until bronchospasm is controlled. Similarly, a wheezing patient using accessory muscles of respiration may not tolerate the mid-to-high thoracic motor blockade induced by spinal or epidural anesthesia and may require intubation and general anesthesia to support ventilation.
OBSTRUCTIVE SLEEP APNEA
Sleep apnea is of particular concern to anesthesiologists.63 Sleep apnea can be either obstructive or central. Normal sleep is a state of rousable unconsciousness. Arousal protects apneic patients from further hypoxemia and hypercarbia. Residual anesthetic-induced or analgesic-induced sedation interferes with normal arousal, with potentially life-threatening consequences in patients with sleep apnea. Although not rigorously studied, residual anesthetic agents, sedatives, and analgesics appear to worsen or precipitate obstructive sleep apnea by decreasing pharyngeal tone, depressing reflex ventilatory responses to hypoxia and hypercarbia, and inhibiting arousal responses to hypoxia and hypercarbia. Central respiratory depression may also result.63 Aggravating factors include the site of surgery (e.g., upper abdominal, thoracic, or upper airway procedures), and nursing in the supine position.
Symptoms of sleep apnea include heavy snoring, sudden nighttime awakenings, witnessed apnea by a bed partner, and daytime somnolence and headache. In as much as sleep symptoms are common during pregnancy, the diagnosis of sleep apnea may not be entertained.64 The incidence of obstructive sleep apnea syndrome is likely increased during pregnancy.65 Obese pregnant women are at increased risk for sleep-related disordered breathing.66 Patients with diagnosed sleep apnea, or a history suggestive of sleep apnea, should be considered at high risk for difficult airway management under anesthesia.
Anesthetic and postoperative analgesic techniques that minimize the use of sedatives and opioid analgesics are safer for these patients.63 Usually, this involves neuraxial anesthesia/analgesia or some type of nerve block. The intensity of postoperative monitoring and nursing care depends on the severity of sleep apnea, the anesthetic and surgical procedure, and need for postoperative opioid analgesia. Patients who use continuous positive airway pressure (CPAP) preoperatively should continue the therapy in the immediate postoperative period (starting in the postanesthesia care unit). Oxygen therapy alone is not adequate.
The diagnosis and treatment of congenital heart disease has improved markedly over the past several decades, leading to an increase in the number of pregnant women with congenital heart disease. Women may present in pregnancy with corrected, partially corrected, or uncorrected lesions. A referral for an anesthesiology consult prior to delivery is indicated. A multidisciplinary team approach involving the obstetrician, cardiologist, and anesthesiologist may provide the best maternal and fetal outcomes.
Pregnant women with corrected lesions and normal heart function may require no special interventions, except peripartum antibiotic prophylaxis if indicated. Women with lesions resulting in left-to-right intracardiac shunting, for example, small atrial septal, or ventricular septal defects, or a patent ductus, should consider early neuraxial analgesia.67 Early administration of neuraxial analgesia will mitigate the pain-induced increased in systemic vascular resistance (SVR), which might worsen the left-to-right shunt and result in pulmonary hypertension or right heart failure. Infusion of air in both intravenous and epidural catheters should be avoided, as a transient right-to-left shunt could result in a paradoxical air embolus. Epidural analgesia should be induced slowly to avoid an acute decrease in SVR that might cause a right-to-left shunt and maternal hypoxemia. Alternatively, neuraxial analgesia can be induced with intrathecal opioids, which avoids the acute onset of local anesthetic-induced sympathectomy and decrease in SVR. Finally, maternal oxygen saturation should be monitored, because mild hypoxemia can lead to an increase in pulmonary vascular resistance and a right-to-left shunt. Similarly, hypercarbia and acidosis can also worsen pulmonary hypertension and lead to shunt reversal.
Pregnant women with cyanotic heart lesions (right-to-left shunts) require special care. Neuraxial anesthesia-induced decreases in SVR may acutely worsen the shunt and hypoxia, which in turn worsen pulmonary hypertension. Examples of these lesions include uncorrected tetralogy of Fallot and Eisenmenger's syndrome. The anesthetic management goals in Eisenmenger's syndrome include maintaining SVR, intravascular volume and preload (traditional epidural analgesia does the opposite), prevent pain, hypoxia, hypercarbia and acidosis, which worsen pulmonary vascular resistance, and avoid myocardial depression (with anesthetic agents).67,68,69 Supplemental oxygen, maternal pulse oximetry, and perhaps invasive arterial blood pressure monitoring are indicated. Pulmonary artery catheter monitoring is controversial, and the risks may outweigh the benefits.67,70,71,72 For vaginal delivery, intrathecal opioids can provide analgesia during the first stage of labor without an associated sympathectomy. Second stage analgesia can be provided with a pudendal block or the judicious use of local anesthetics administered into the lumbar epidural space.67 Both general and epidural anesthesia have been used for cesarean delivery.67 The hemodynamic changes that occur in the immediate postpartum period are especially stressful to the patient with Eisenmenger's syndrome.73
The anesthetic considerations in caring for patients with primary pulmonary hypertension mimic those of Eisenmenger's syndrome. Single-shot spinal anesthesia is contraindicated for both these lesions, although slow-onset epidural anesthesia has been used for cesarean deliveries.74
The anesthetic considerations in parturients with asymmetric septal hypertrophy include maintaining intravascular volume and venous return, maintaining SVR, maintaining a slow heart rate and sinus rhythm, and preventing an increase in myocardial contractility.67 Beta-adrenergic receptor blockade should be continued during labor and delivery. Neuraxial analgesia techniques associated with a slow onset of sympathectomy, as discussed, are appropriate for these patients. Phenylephrine, rather than ephedrine or epinephrine, is the preferred vasopressor. Although slow-onset epidural anesthetic has been described for cesarean deliveries, these patients tolerate general anesthesia well.75 Single-shot spinal anesthesia is contraindicated.
Women with valvular heart lesions also require special consideration. In general, stenotic lesions are less well-tolerated than regurgitant lesions. The anesthetic management goals of aortic stenosis include maintaining normal heart rate and sinus rhythm, maintaining SVR, maintaining intravascular volume and venous return, and avoiding drug-induced myocardial depression.67,76 Moderate-to-severe aortic stenosis is an indication for peripartum invasive arterial pressure monitoring. A central venous or pulmonary artery catheter may be used to monitor volume status. Single-shot spinal anesthesia is contraindicated. Neuraxial labor analgesic techniques that avoid an acute decrease in preload, as described, are appropriate. Slow-onset epidural anesthesia has been described for cesarean delivery,76,77 although some anesthesiologists prefer general anesthesia for women with severe aortic stenosis.
The anesthetic goals for patients with mitral stenosis are similar to aortic stenosis. Sinus rhythm should be maintained, if present, and rapid atrial fibrillation should be treated aggressively. A slow heart rate improves ventricular filling and cardiac output, as does the avoidance of hypovolemia. However, hypervolemia may lead to pulmonary edema. Finally, pain, hypoxemia, hypercarbia, and acidosis may increase pulmonary vascular resistance. Again, a neuraxial labor technique that insures the slow onset of sympathectomy is appropriate. Small doses of phenylephrine should be used to treat hypotension. Tachycardia can be treated with esmolol. After delivery, oxytocin, methylergonovine, and 15-methyl prostaglandin F2a may increase pulmonary vascular resistance.67 Finally, the patient with mitral stenosis is most at risk for pulmonary edema in the immediate postpartum period, as peripartum cardiac output is the highest during this time. These patients may benefit from the sympathectomy and analgesia afforded by continuing a continuous infusion of epidural local anesthetics in the immediate postpartum period.67 Intensive care may be required.
Neuraxial analgesia/anesthesia is beneficial to women with aortic insufficiency. It prevents an increase in SVR, which worsens aortic insufficiency. Other hemodynamic goals include maintaining a normal or slightly increased heart rate, maintaining preload, and avoiding drug-induced myocardial depression.67
Similarly, women with mitral regurgitation benefit from the decrease in SVR associated with neuraxial analgesia/anesthesia. Normal sinus rhythm should be maintained, with a normal or slightly increased heart rate. Preload should be maintained, but hypervolemia should be avoided, as should increases in pulmonary vascular resistance. Invasive hemodynamic monitoring may be warranted, as well as continuous ECG monitoring.
Women with mitral value prolapse are good candidates for neuraxial analgesia/anesthesia. Phenylephrine, rather than ephedrine, may be preferable for treating hypotension to avoid precipitating tachydysrhythmia.
Women with prosthetic heart values are often anticoagulated with heparin or low-molecular-weight heparin. Neuraxial analgesia/anesthesia is contraindicated when anticoagulation is maintained during labor and delivery. The American Society of Regional Anesthesia and Pain Medicine has recently published an updated consensus statement regarding neuraxial anesthesia and anti-coagulation (http://www.asra.com/items_of_interest/consensus_statements/index.iphtml). In general, neuraxial analgesia/anesthesia should not be initiated for 24 hours after the last therapeutic dose of low-molecular-weight heparin. Low-molecular-weight heparin should not be administered for at least 2 hours after epidural catheter removal. Regular heparin has a shorter half-life than low-molecular-weight heparin, but chronic use can be associated with thrombocytopenia; therefore, a platelet count should be checked before initiating neuraxial blockade. Aspirin use does not contraindicate neuraxial blockade.
There are several considerations when caring for the pregnant patient with a transplanted heart.67 The immunosuppressive regimen may include corticosteroids, and stress doses of steroids should be administered in the peripartum period. Aortocaval compression and maintenance of intravascular volume is essential, because the heart does not reflexly compensate for hypovolemia. There is up-regulation of cardiac beta-adrenergic receptors, resulting in increased sensitivity of adrenergic agents (e.g., epinephrine in local anesthetic solutions78). Neuraxial blockade should be induced slowly to decrease the risk of hypotension. Additional intravenous fluids and direct acting vasopressors can be used to treat hypotension.79
Regional labor analgesia is beneficial to patients with ischemic heart disease. Effective analgesia decreases maternal circulating catecholamine levels,57,80 thus preventing tachycardia and increases in myocardial work. Patients with peripartum cardiomyopathy, or other myopathies, also benefit from neuraxial blockade.81 This avoids the myocardial depressant effects of systemic anesthetic agents.
SICKLE CELL DISEASE
Epidural labor analgesia may aid in the care of the laboring patient with sickle cell disease in several ways.82 Pain control is of obvious importance in the patient with sickle cell disease. Epidural analgesia to manage pain during a sickle crisis that occurred during labor has been described.83 An additional advantage is that epidural blockade is associated with decreased peripheral venous stasis.84 Euvolemia should be maintained, as well as normothermia.
Regional anesthesia is contraindicated in the presence of frank coagulopathy. Spinal/epidural hematoma may result in permanent neurologic damage if not diagnosed and decompressed within 6 to 12 hours after onset of symptoms.85 Unfortunately, laboratory tests of coagulation are often neither sensitive nor specific for clinical bleeding. A clinical history of bleeding and physical examination looking for bruising or bleeding from venipuncture sites may be a better indicator of the risk of bleeding. However, pregnant women often have epistaxis and bleeding around the gums without having a coagulopathy.
There are no data to support a specific platelet count cut-off, above which neuraxial analgesia/anesthesia is safe. Epidural or spinal hematoma is a rare complication of neuraxial anesthesia in pregnant women85 and is seen most often in patients with a known coagulopathy.86 Almost all anesthesiologists will initiate neuraxial anesthesia if the platelet count is more than 100,000/mm3. A platelet count of 50,000/mm3 is the absolute lower limit acceptable to most anesthesiologists.87 These same parameters apply for removing an epidural catheter as epidural hematomas can occur after catheter removal in a coagulopathic patient.85
Patients with hypercoagulable states, for example, protein C, S, or antithrombin III deficiency, or Leiden factor V, may be receiving either prophylactic or therapeutic doses of heparin. Prophylactic regular heparin (5000 units subcutaneous) is not a contraindication to neuraxial blockade, but 12 hours should elapse after the last dose of prophylactic enoxaparin, and 24 hours after dalteparin. The Consensus Statement from the American Society of Regional Anesthesia and Pain Medicine is discussed.
The decision to proceed with neuraxial analgesia in patients with borderline coagulation status, or unknown coagulation status, should involve an assessment of risk versus benefit. Labor epidural analgesia is an elective procedure for most women. The risk of epidural hematoma is greater when large-bore epidural needles and catheters are used compared with a small-gauge single-shot spinal needle.88 Women with continuous epidural analgesia may have sensory blockade for many hours and therefore may not perceive the symptoms of a developing spinal/epidural hematoma (progressive back and leg pain, urinary retention, and lower extremity weakness). In contrast, the benefit may outweigh the risk for women who are at high risk for cesarean delivery and difficult intubation, for example, women with preeclampsia, because the risk of difficult intubation is probably greater than the risk of spinal/epidural hematoma. Use of continuous epidural infusion of dilute solutions of local anesthetic, combined with opioid, minimize motor blockade during continuous epidural labor analgesia, and may be safer because there is less risk of masking the symptoms of hematoma. If the decision is made to proceed with neuraxial analgesia/anesthesia, regular lower extremity neurologic checks should be performed during and after neuraxial blockade.
Epidural analgesia/anesthesia is more likely to be incomplete or fail in patients with chronic back pain and previous back surgery.89,90,91 Onset of epidural analgesia can be delayed in patients with back pain or disc disease.89 Epidurography demonstrated that contrast material injected into the epidural space did not reach the nerve root in 33% of patients with disc herniation and did not move beyond the affected interspace in 5% of patients.92 In one study, epidural anesthesia was successful in 91% of patients with previous limited back surgery, compared with 98.7% in patients with no history of back surgery.90 Lumbar epidural labor analgesia is usually initiated in the lower lumbar interspaces. Afferent stimuli during the first stage of labor originate from the L1 through T10 dermatomes, whereas stimuli from the late first stage and second stage of labor also originate from sacral dermatomes. Therefore, anesthetic injected into the epidural space above the site of lumbar spinal surgery may diffuse upwards, but sacral analgesia may be incomplete. Conversely, anesthetic injected below the level of surgery may provide satisfactory sacral analgesia, but first stage analgesia may be incomplete. A two-catheter technique has been described.93 Spinal analgesia/anesthesia is likely to be more successful than epidural analgesia/anesthesia. It is helpful to refer these patients for anesthesia evaluation before delivery, or at least to ask the patient to obtain a copy of her surgical report. The back skin scar is often more extensive than the actual surgery.
Patients with scoliosis should also be referred for evaluation before delivery. Review of spine radiographs and operative reports help determine the level of fusion and identify interspaces that are less distorted. These patients may be candidates for neuraxial analgesia/anesthesia. However, it is technically more difficult to identify the epidural space, multiple attempts may be necessary, and the incidence of unintentional dural puncture is higher, as is the incidence of failed or incomplete block.94,95 Continuous caudal analgesia has been used successfully for patients with spine instrumentation.96
Malignant hyperthermia is a hypermetabolic disease of skeletal muscle that is triggered in genetically susceptible people by exposure to volatile anesthetic agents or succinylcholine. It is life-threatening if a crisis is not recognized and treated appropriately. A history of malignant hyperthermia susceptibility in the parturient or her family should prompt an antenatal interview with an anesthesiologist. Additionally, the anesthesia team should be notified when the patient is admitted for labor and delivery, because the anesthesia machine and drugs need to be specially prepared. Epidural analgesia for labor will decrease the likelihood that a patient with malignant hyperthermia susceptibility will require general anesthesia for emergency cesarean delivery. Although it is possible to administer safe general anesthesia to the malignant hyperthermia-susceptible patient, it is more cumbersome in the emergency situation, because the drugs commonly used to induce general anesthesia for emergency cesarean delivery are also triggers of malignant hyperthermia.
Neuraxial analgesia/anesthesia is contraindicated in the presence of systemic infection (risk of hematogenous spread if the procedure is performed during bacteremia or viremia) or local infection at the site of skin puncture. Reports of spinal/epidural abscess or meningitis in obstetric patients are rare.97 Symptoms of spinal/epidural abscess may take days or months to develop.98,99 Fever and backache are followed by radicular symptoms (sensory deficits and motor and sphincter weakness over a span of hours to weeks98). Leukocytosis is usually present. Similarly, the onset of meningitis can also vary from days to weeks.97
Human immunodeficiency viral (HIV) particles can be isolated from cerebrospinal fluid (CSF) at the time of initial infection.100 Therefore, neuraxial analgesia/anesthesia does not risk introduction of the virus into the central nervous system, because this has already occurred. In a group of 18 HIV-positive patients who received neuraxial anesthesia, there was no evidence of accelerated disease progression or increased infectious or neurologic complications after delivery.101 Similarly, in a group of 45 HIV-infected women who were receiving antiviral therapy, and who underwent elective cesarean delivery under spinal anesthesia, there were no intraoperative or postoperative complications compared with a cohort of uninfected women.102
Some HIV patients may use protease inhibitors. This class of drugs is metabolized in the liver by the cytochrome P-450 system, specifically by the CYP3A isoenzyme.103 The drugs also competitively inhibit this enzyme. Many sedative-hypnotic and analgesia agents are also metabolized by the CYP3A isoenzyme, for example, midazolam, meperidine, fentanyl, and methylergonovine. These drugs are relatively contraindicated in patients using protease inhibitors.104 However, clinical experience suggests they can be used, but careful titration and monitoring is necessary.105
Patients with symptomatic herpes simplex virus type 2 (HSV-2) often present for cesarean delivery. Primary HSV-2 infection is associated with transient viremia, but recurrent infections are not.106 Retrospective studies have found no sequelae after neuraxial anesthesia in patients with symptomatic HSV-2.107,108,109,110 However, very few patients in these studies had primary infections. There is a general consensus among anesthesiologists that neuraxial anesthesia is safe for parturients with secondary HSV-2 infections. There are not enough data to support the safety of neuraxial anesthesia in parturients with primary infections. Therefore, the theoretical risk of introducing virus into the central nervous system must be weighed against the risks of general anesthesia.111
Many drugs, both legal and illegal, have the potential to adversely impact the anesthetic care of a patient.
There are increased risks to both general and neuraxial anesthesia in cocaine-abusing parturients. General anesthesia is associated with a higher risk of severe hypertension and dysrhythmias.112 In an animal model, acute cocaine intoxication and general anesthesia also resulted in myocardial ischemia.113 Long-term cocaine use is associated with a increased sensitivity to inhalation anesthetic agents, whereas acute toxicity may increase the requirement for anesthetic agents.114
There are several considerations when choosing neuraxial analgesia/anesthesia. Controversy exists as to whether cocaine abuse is associated with thrombocytopenia.112,115 After neuraxial blockade, the degree of hypotension is more profound.112,115 Because ephedrine is both an indirect- and direct-acting vasopressor, it is less effective in treating hypotension in the cocaine-positive parturient.112 Finally, cocaine-abusing parturients often report pain despite apparent adequate regional anesthesia.114
Acute and/or long-term amphetamine abuse complicates the administration of both general and neuraxial anesthesia. As with cocaine, general anesthesia is associated with hypertension and dysrhythmias.116 Acute ingestion increases anesthetic requirements, while long-term use decreases anesthetic requirements.117 Neuraxial anesthesia may be associated with severe hypotension and unpredictable response to vasopressors. There are two reported cases of cardiac arrest after general118 and regional119 anesthesia for cesarean delivery.
Long-term opioid use affects the administration of analgesia/anesthesia in several ways. The parturient's baseline opioid use should be calculated and administered to prevent maternal and neonatal withdrawal. Treatment of labor pain, or postoperative pain, will require additional doses of opioid. Neuraxial analgesic/anesthetic techniques may be advantageous for the opioid addict because analgesia can be provided with local anesthetics (without opioids). As with cocaine abuse, apparently adequate regional anesthesia may not provide acceptable pain relief in some opioid users.120 Finally, intravenous drug users have an increased risk of spontaneous epidural abscess.121
Cannabis may enhance the sedative–hypnotic effects of other central nervous system depressants.122 It impairs lung function in a manner similar to tobacco. Acute cannabis intoxication has been associated with uvular edema and airway obstruction after general anesthesia.123
There are few studies of the anesthetic management of diabetic parturients. In nonpregnant diabetic patients with autonomic neuropathy, general anesthesia is associated with higher vasopressor use.124 Gastroparesis, another manifestation of autonomic dysfunction, may increase the risk of pulmonary aspiration.
Labor and delivery represents one of the few situations in which an anesthetic might have to be administered to a patient who is not euthyroid. Patients should be made euthyroid if at all possible before a surgical procedure or if there is a risk of perioperative thyroid storm. No prospective randomized studies have compared the safety of anesthetic techniques in hyperthyroid patients.125 Potential anesthetic considerations include the hyperdynamic cardiovascular system and possibility of cardiomyopathy, airway obstruction from an enlarged thyroid gland, respiratory muscle weakness, and electrolyte abnormalities.125 The hyperthyroid patient who requires emergency surgery should receive propylthiouracil, sodium iodide, glucocorticoids, and propranolol.125,126
Clinical manifestations of hypothyroidism that may potentially affect anesthetic management include myocardial dysfunction, coronary artery disease, defects in respiratory drive, obstructive sleep apnea, paresthesias, hyponatremia, decreased glucocorticoid reserves, anemia and abnormal coagulation factors, and platelet dysfunction.125 Again, no studies have addressed whether one type of anesthetic management is superior to another in these patients. The coagulation system should be assessed before initiation of neuraxial analgesia/anesthesia.
Pregnant women with pheochromocytoma may require tumor resection early in pregnancy or cesarean delivery followed by tumor resection.127 General, neuraxial, and combined anesthetic techniques have been used successfully for tumor resection,128,129,130 but no randomized study has been performed.
NEUROLOGIC AND NEUROMUSCULAR DISORDERS
There are several anesthetic considerations for women with epilepsy. Many anticonvulsants induce hepatic liver enzymes and affect the metabolism of other drugs, including drugs commonly used in the administration of anesthesia.131 In addition, these women may be at risk for deficiency in their vitamin K-dependent clotting factors.132 Several anesthetic drugs decrease the seizure threshold and should be avoided. These include ketamine, enflurane, and meperidine.133
There is controversy as to the appropriate anesthetic/analgesic for the laboring patient with multiple sclerosis. There is theoretic concern that exposing demyelinated areas of the spinal cord to local anesthetic agents could result in worsening of the disease. Two small retrospective reports identified a small percentage of patients with multiple sclerosis who had a relapse after spinal anesthesia.134,135 However, this can also occur after general anesthesia. In a retrospective series of 32 parturients with multiple sclerosis who received neuraxial analgesia/anesthesia, the incidence of relapse was not greater than in a cohort of women who received local infiltration analgesia.136 The pregnant woman with multiple sclerosis should be counseled that there is a increase in exacerbations within the first several months after delivery137 and that it is not known whether neuraxial anesthesia influences the risk of a relapse.
Myasthenic gravis may be worsened by a number of drugs, including drugs commonly used in the provision of anesthesia. Patients with myasthenia gravis are extremely sensitive to nondepolarizing muscle relaxants.138 Their response to succinylcholine is unpredictable.139 Neuraxial analgesia/anesthesia is the anesthetic technique of choice in these patients because it avoids the potential respiratory depressant effects of systemic opioids and makes the need for general anesthesia less likely. Patients with myasthenia gravis are at increased risk for requiring postoperative mechanical ventilation.138
Women with spinal cord injuries above the level of T7 to T5 are at risk for autonomic hyperreflexia during labor and delivery. Neuraxial anesthesia with local anesthetic agents will prevent labor-induced autonomic hyperreflexia. Epidural fentanyl without local anesthetic did not prevent autonomic hyperreflexia during labor.140 An additional anesthetic concern in patients with spinal cord injury is the use of succinylcholine. Succinylcholine used within 1 year of the injury can precipitate life-threatening hyperkalemia.141
The anesthetic management of parturients with liver disease, who have no hepatic synthetic or metabolic impairment, is unchanged from the routine. However, parturients with acute viral hepatitis, advanced cirrhosis, portal hypertension, or other diseases that are associated with liver dysfunction often have multiorgan disease as a result of their liver dysfunction. Liver parenchymal disease may result in deficiencies of clotting factors synthesized in the liver. Cholestasis can lead to malabsorption of vitamin K and deficiencies in the vitamin K-dependent clotting factors.142 Therefore, coagulation status should be assessed before the initiation of neuraxial analgesia/anesthesia.
Cardiovascular changes associated with liver disease include increased cardiac output and low systemic vascular resistance. Blood volume is increased and there may be a cardiomyopathy. Invasive monitoring may be indicated. Portopulmonary shunts and impaired hypoxic pulmonary vasoconstriction contribute to hypoxia. Orogastric or nasogastric tubes are relatively contraindicated in the presence of esophageal varices. Other organ systems affected include the brain (hepatic encephalopathy), multiple metabolic derangements, and renal dysfunction.
Ascites may impair venous return, elevate the diaphragm, leading to a more profound decrease in functional residual capacity, and increase intraabdominal pressure, thus increasing the risk of gastroesophageal reflux and pulmonary aspiration. It may also lead to engorged epidural veins, thus complicating epidural anesthesia.
Drug distribution and clearance are altered in liver disease. For example, the volume of distribution of lidocaine is doubled and the half-life increases three-fold in patients with cirrhosis.143
Volatile anesthetic agents all reduce hepatic blood flow.144 It is unclear whether this contributes to postoperative liver dysfunction, or worsening of liver dysfunction. Neuraxial anesthesia also reduces hepatic blood flow.145,146
The parturient with mild-to-moderate renal insufficiency and well-controlled hypertension and euvolemia requires minimal special consideration.147 The half-lives of drugs excreted by the kidney may be prolonged and protein binding is altered. Many organ systems are affected by severe chronic or acute renal insufficiency. Uremic patients may have platelet dysfunction and thrombocytopenia. They may be hypervolemia (before dialysis) or hypovolemia (after dialysis). The slow induction of neuraxial analgesia/anesthesia is indicated to prevent profound hypotension.148 Patients with chronic uremia have delayed gastric emptying and increased gastric acidity, which may increase the risk of pulmonary aspiration.147 Serum potassium should be checked before the administration of succinylcholine, because succinylcholine is associated with a .5- to .7-milliequivalent increase in potassium concentration.149 Intravenous fluids that do not contain potassium should be used.
Patients who have undergone a renal transplantation, without other systemic disease, do not require special anesthetic considerations. Patients with urolithiasis may benefit from epidural analgesia, as this provides analgesia with less fetal exposure to drugs and may facilitate passage of the calculus through decreased ureteral spasm.150
There are a number of anesthetic issues that should be considered when providing anesthetic care to the parturient with chronic disease. In addition, appropriate labor analgesia and surgical anesthesia may help mitigate some of the adverse affects of labor and delivery on chronic disease. A multidisciplinary team approach involving the obstetrician, anesthesiologist, disease specialist, and other health care providers is likely to provide the best outcome. Antepartum consultation allows all members of the team, and the patient, to understand the issues involved and plan for a safe labor and delivery.
25. Cruz AM, Southerland LC, Duke T et al: Intraabdominal carbon dioxide insufflation in the pregnant ewe. Uterine blood flow, intraamniotic pressure, and cardiopulmonary effects Anesthesiology 85:1395, 1996
42. Tame JD, Abrams LM, Ding XY et al: Level of postoperative analgesia is a critical factor in regulation of myometrial contractility after laparotomy in the pregnant baboon: Implications for human fetal surgery. Am J Obstet Gynecol 180:1196, 1999
52. Young TK, Woodmansee B: Factors that are associated with cesarean delivery in a large private practice: The importance of prepregnancy body mass index and weight gain. Am J Obstet Gynecol 187:312, 2002
62. Laszlo A, Buljubasic N, Zsolnai B et al: Interactive effects of volatile anesthetics, verapamil, and ryanodine on contractility and calcium homeostasis of isolated pregnant rat myometrium. Am J Obstet Gynecol 167:804, 1992
104. Raffanti S, Haas DW: Antimicrobial agents: Antiretroviral agents. In: Hardman JG, Limbird LE, (eds): Goodman & Gilman's The Pharmacologic Basis of Therapeutics. pp 1349-1380, 10th ed. New York, McGraw-Hill, 2001
108. Crosby ET, Halpern SH, Rolbin SH: Epidural anaesthesia for caesarean section in patients with active recurrent genital herpes simplex infections: a retrospective review. Can J Anaesth 36:701, 1989
113. Boylan JF, Cheng DC, Sandler AN et al: Cocaine toxicity and isoflurane anesthesia: Hemodynamic, myocardial metabolic, and regional blood flow effects in swine. J Cardiothorac Vasc Anesth 10:772, 1996
144. Frink EJ Jr., Morgan SE, Coetzee A et al: The effects of sevoflurane, halothane, enflurane, and isoflurane on hepatic blood flow and oxygenation in chronically instrumented greyhound dogs. Anesthesiology 76:85, 1992