The effects of multiple sclerosis and long-term use of disease-modifying therapies on fertility are poorly studied. A prospective study in Finland showed that women with multiple sclerosis were more likely to need artificial insemination to initiate pregnancy than women in general. A large Danish national cohort study found that multiple sclerosis was associated with having no or fewer children ; however, it is not known whether this is caused by the disease itself or by altered reproductive behaviors among diseased individuals. In one study of women with multiple sclerosis undergoing assisted reproductive technology, failed assisted reproductive attempts were associated with increased annual relapse rates. These findings may be explained by the stress of infertility, temporary interruption in disease-modifying therapy, and increases in inflammatory cytokines associated with hormonal therapies.

Pregnancy and obstetric outcomes in women with multiple sclerosis are likely no different from individuals without disease. One cohort study compared 198 affected women with 1584 healthy women; the number of maternal complications was not higher in women with multiple sclerosis. However, infants of women with multiple sclerosis were at greater risk for meconium aspiration. This finding may reflect an intrauterine environment in patients with multiple sclerosis that is more susceptible to acute hypoxic events. A subsequent cohort study of 649 pregnancies in women with multiple sclerosis concluded that infants of these women were more likely to be small for gestational age, attributed to a suboptimal intrauterine environment. Moreover, this study found that mothers with multiple sclerosis were more likely to undergo induction of labor and operative delivery, possibly as a result of neuromuscular weakness and spasticity. These findings have not been reproduced in other studies. A 2017 UK case-controlled study of 181 pregnancies in 98 mothers with multiple sclerosis and 244,573 pregnancies in 124,830 mothers without multiple sclerosis did not find any associations with neonatal birth weight, gestational age at birth, mode of delivery, stillbirth, or neonatal death. A 2011 metaanalysis did not find significant associations between multiple sclerosis and adverse obstetric and neonatal outcomes. The relapse rate was lower during pregnancy than before or after pregnancy.

Data from prospective studies suggest that the rate of relapse increases during the first 3 months postpartum compared with the year before pregnancy. Relapses during this period were more likely in women who had higher relapse rates in the year before pregnancy or during pregnancy. Stress, exhaustion, infection, the loss of antenatal immunosuppression, and the postpartum decline in concentrations of reproductive hormones may account for the higher postpartum relapse rate. Treatment with immunologically active agents (e.g., interferon-beta) may result in a decreased postpartum relapse rate, but data are limited.

Pregnancy does not negatively affect the long-term outcome of multiple sclerosis. Data are conflicting as to whether exclusive breastfeeding is associated with a lower risk for relapse than partial or no breastfeeding. A 2015 prospective study in 201 German women found a modestly protective effect against relapses among women with multiple sclerosis who exclusively breastfed for at least 2 months (first postpartum relapse adjusted hazard ratio, 1.70; 95% confidence interval [CI], 1.02 to 2.85; P =.04).

The anesthesia provider should assess the patient’s level of compromise, document the pattern of deficits, and give special attention to respiratory involvement, in particular the ability to cough, clear secretions, and take vital capacity breaths. Historically, the optimal mode of anesthesia in patients with multiple sclerosis has been controversial. Most anesthesia providers have considered general anesthesia to be safe. Many anesthesia providers have been reluctant to administer neuraxial anesthesia because the effect of local anesthetic drugs on the course of the disease is unclear. Some anesthesia providers have expressed concern that neuraxial anesthesia may expose demyelinated areas of the spinal cord to potentially neurotoxic effects of local anesthetic agents. Several animal studies have investigated the histologic effects of local anesthetic agents on the normal spinal cord. In one study, subarachnoid injection of small doses of a local anesthetic agent produced no histologic changes in the spinal cord or meninges. Injection of very large doses caused reversible inflammatory and degenerative changes, but all changes resolved within 14 days of injection.

Diagnostic lumbar puncture is not associated with a higher rate of relapse. Historically, two small reports implicated spinal anesthesia in the exacerbation of multiple sclerosis. Bamford et al. described one case of relapse after the administration of spinal anesthesia in nine patients, and Stenuit and Marchand identified two cases of relapse after the administration of spinal anesthesia in 19 patients. Notably, there was no evidence to suggest that these relapses were related to spinal anesthesia, rather than to other factors known to exacerbate multiple sclerosis such as postoperative conditions (e.g., stress, infection, hyperpyrexia) or simply the postpartum period. Two larger observational studies have not found an association between neuraxial analgesia and anesthesia and an increased risk of postpartum relapse. One observational cohort study found that postpartum relapse was primarily a function of disease activity, but was not associated with neuraxial procedures during labor and delivery.

Available data on relapse outcomes associated with epidural analgesia and anesthesia in patients with multiple sclerosis show some cases of relapse, as would be expected. Bader et al. retrospectively evaluated 32 pregnancies in women with multiple sclerosis; they observed that women who received epidural anesthesia for vaginal delivery did not have a higher incidence of relapse than those who received only local infiltration anesthesia. The Pregnancy in Multiple Sclerosis study followed 227 women who had multiple sclerosis for at least 1 year before conception, of whom 42 received epidural analgesia during labor; no adverse effect of epidural analgesia on the rate of postpartum relapse or the progression of disability was identified. Similarly, a 2012 prospective study that followed 349 patients for 5 years postpartum did not find an increased risk for relapse associated with labor epidural analgesia. A 2013 record linkage study from British Columbia compared 431 deliveries in women with multiple sclerosis who received peripartum epidural analgesia/anesthesia versus 2959 deliveries in women from the general population, and 128 women with multiple sclerosis who received spinal anesthesia for cesarean delivery versus 846 women in the general population; neither neuraxial technique was associated with increased disability.

Although there are limited data on spinal and general anesthesia for cesarean delivery in patients with multiple sclerosis, they are both considered safe, and prior concerns over these techniques are attributable to recall bias associated with postpartum relapses after regional anesthesia. No data suggest harmful effects of neuraxial opioids in women with multiple sclerosis. Considering the significant benefits of neuraxial techniques for intraoperative anesthesia and postoperative analgesia, either spinal or epidural anesthesia is the principal anesthetic technique used for cesarean delivery in patients with multiple sclerosis in many institutions.

In summary, published data do not contraindicate the use of neuraxial anesthetic techniques for labor analgesia or operative anesthesia. The patient should be aware that there is a higher incidence of relapse during the postpartum period, even without the use of neuraxial analgesia or anesthesia. In addition, when anesthetic techniques are used, the type of anesthesia selected does not appear to influence the relapse rate. Neither pregnancy nor anesthesia appears to have a negative influence on the long-term course of the disease. The willingness of anesthesia providers to use neuraxial techniques in pregnant patients with multiple sclerosis is reflected in a survey of obstetric anesthesiologists in the United Kingdom published in 2006. The majority (91%) of respondents had seen fewer than 10 cases of multiple sclerosis in the past 10 years; 79% and 98% of anesthesia providers indicated they would perform a neuraxial anesthetic technique for labor and elective cesarean delivery, respectively.

Tension or muscle contraction headaches are the most common type of headache observed during pregnancy. The symptoms typically consist of dull, persistent pain that extends over the entire head. The onset is usually gradual, but the symptoms may persist for long periods. Although the etiology is unknown, this type of headache is believed to be associated with stress rather than hormonal changes. These headaches are more common in women, are frequently associated with anxiety, and may be a symptom of postpartum depression.

A migraine headache is classically described as a unilateral, throbbing headache sometimes accompanied by nausea and vomiting. The duration varies from hours to days. Visual disturbances (e.g., scotomata) typically precede onset, and focal neurologic symptoms (e.g., aphasia, hemiplegia) may also occur. Most neurologists favor neurovascular vasospasm, followed by cerebral vasodilation, as a cause of these headaches; a primary vascular disorder or a disturbance in the noradrenergic nervous system also may be involved. Patients appear to be more susceptible to symptoms when serotonin levels are low.

The 1-year period prevalence of migraine headaches in the United States is 3.9% for men and 5.1% for women. Prevalence is higher in middle life, between 30 and 59 years of age. Hormonal influences have a strong association with migraine headaches; estrogen withdrawal is associated with an exacerbation of symptoms. After delivery, the reduction in hormonal concentrations coincides with an increase in migraine symptoms.

Reversible cerebral vasoconstriction syndrome (RCVS) , characterized by acute and severe headache secondary to multifocal cerebral artery vasoconstriction, is often misclassified as a migraine. An estimated 10% of RCVS cases occur in the peripartum period and have been associated with preeclampsia. Diagnosis is by magnetic resonance angiography with digital subtraction, and treatment involves calcium channel blockers , aspirin , and lifelong monitoring for future risk for stroke.

Although there is no estimate of the global incidence of pregnancy in women with spinal cord injury, the World Health Organization (WHO) estimates that nearly 18% of spinal cord injuries worldwide occur in women of reproductive age. Pregnancy may aggravate many of the medical complications of spinal cord injury ( Box 47.1 ). The decreases in both functional residual capacity and expiratory reserve volume during pregnancy may increase the likelihood of respiratory compromise associated with spinal cord injury, with worsening of respiratory function occurring as early as 20 weeks’ gestation. Patients may require tracheal intubation and mechanical ventilatory support, and cesarean delivery may be indicated to avert the fetal risks associated with maternal hypercapnia and to improve maternal respiratory mechanics. Complications of acute spinal cord injury in pregnancy include preterm birth and neonatal intensive care unit admission.

Pregnancy increases the risks for thromboembolic complications and urinary tract infection. Loss of sympathetic tone below the level of the lesion renders pregnant patients with spinal cord injury particularly prone to orthostatic hypotension, which may result in reduced uteroplacental perfusion, necessitating careful monitoring of the fetal heart rate (FHR). In pregnant women, autonomic hyperreflexia occurs most commonly during labor; it may be triggered by uterine contractions, vaginal and cervical examinations, speculum insertion, and urethral catheterization. Autonomic hyperreflexia can be distinguished from other causes of intrapartum hypertension by the occurrence of cyclic hypertension (i.e., blood pressure increases during contractions and decreases between contractions). Symptoms of autonomic hyperreflexia are also common in the immediate postpartum period, possibly triggered by postpartum pain, urethral catheterization, and uterine contractions.

Women with a lesion above T11 may be at higher risk for preterm labor. Because these women do not experience labor pain, obstetric management includes weekly cervical examinations during the third trimester, and patients are instructed on uterine palpation techniques to detect contractions at home. Although vaginal delivery is preferred, the development of autonomic hyperreflexia in the second stage of labor may necessitate expedited operative vaginal delivery. Assisted vaginal delivery may also be necessary because of the inability of the mothers to push effectively during the second stage. A case series of 109 births to patients with spinal cord injuries found predominantly favorable maternal and neonatal outcomes, with 37 (34%) undergoing cesarean delivery and 16 (15%) deliveries occurring preterm.

Antenatal anesthetic consultation and multidisciplinary planning are essential to optimize outcomes. The ACOG recommends that a multidisciplinary team in a unit capable of intensive hemodynamic monitoring should deliver parturients with a spinal cord injury. Early neuraxial anesthesia is preferred for the prevention or treatment of autonomic hyperreflexia during labor and delivery. Although neuraxial labor analgesia may attenuate the risk and symptoms of autonomic hyperreflexia, it may not ablate the phenomenon entirely, particularly when low doses of local anesthetic are used. Spinal anesthesia has long been reported as effective for controlling blood pressure in paraplegic patients undergoing general surgical procedures. If spinal anesthesia is chosen, insertion of an intrathecal catheter and use of a continuous technique may be appropriate; this approach may allow careful titration of the resulting neuroblockade. Assessment of the level of blockade in an insensate patient may be accomplished by assessing for loss of lower extremity deep tendon reflexes and meticulous monitoring for acute hypertension and bradycardia.

Most obstetric anesthesia providers prefer the use of epidural analgesia for the prevention or treatment of autonomic hyperreflexia during labor and vaginal delivery. Consideration also should be given to providing epidural analgesia after vaginal delivery to minimize the possibility of autonomic hyperreflexia, which has been reported to occur in response to pain as late as 5 days after delivery. Maintaining optimal analgesia with a transition to oral agents is important.

Case reports have historically described the successful epidural administration of 0.25% or 0.5% bupivacaine, or the administration of combined spinal-epidural (CSE) anesthesia for the mitigation of autonomic hyperreflexia. Baraka reported the successful use of epidural meperidine (pethidine), an opioid with local anesthetic properties, for controlling autonomic hyperreflexia. Abouleish et al. observed that epidural fentanyl alone did not effectively treat the hypertension of autonomic hyperreflexia, but the addition of 0.25% bupivacaine led to a decrease in blood pressure to baseline levels. Maehama et al. described the successful use of intravenous magnesium sulfate for management of autonomic hyperreflexia during labor.

Patients with spinal cord injury often have a low baseline blood pressure and some hemodynamic instability. Placement of an intraarterial catheter before induction of anesthesia allows the continuous assessment of blood pressure and early detection and treatment of symptoms. Pulse oximetry is particularly useful in patients with respiratory compromise, and the anesthesia provider should always be available to provide ventilatory assistance if necessary.

Positioning for neuraxial block may be difficult; the anesthesia provider should consider performing the block with the patient in a lateral position because the sitting position may cause hypotension from venous pooling in the lower body. Therapeutic doses of local anesthetic should be administered cautiously with the understanding that the cephalad level of the sensory block can be fully assessed only if it is higher than the level of the spinal cord lesion. As a result, the typical epidural test dose may not identify unintentional subarachnoid injection in a patient with a spinal cord injury. Neuraxial blockade can be partially assessed by evaluating segmental reflexes below the level of the lesion. For example, the anesthesia provider can lightly stroke each side of the abdomen above and below the umbilicus, looking for contraction of the abdominal muscles and deviation of the umbilicus toward the stimulus. Reflexes are absent below the level of the block. In some patients with spastic paresis at baseline, the level of anesthesia may be confirmed by the conversion of spastic paresis to flaccid paresis. A decline in blood pressure may also herald the onset of neuraxial blockade.

Alternative means of treating autonomic hyperreflexia should be available at the bedside if neuraxial analgesia or anesthesia is inadequate or not successful. Antihypertensive medications, such as nitroglycerin , or other vasodilators, may be effective, recognizing that hypotension can result in decreased uterine blood flow. Careful titration of sodium nitroprusside , noting the potential for fetal/neonatal cyanide intoxication, or beta-adrenergic receptor blockade , may also be useful. The anesthesia provider should recognize that increased vagal activity during autonomic hyperreflexia can result in electrocardiographic changes including first- and second-degree atrioventricular block and sinus arrest.

If cesarean delivery is necessary, spinal, epidural, or CSE anesthesia can be administered. Spinal anesthesia is associated with more rapid onset and more frequent need to treat hypotension. The effect of neuraxial blockade on respiratory function may be less severe with epidural anesthesia than with spinal anesthesia.

Severe respiratory insufficiency or technical difficulties with neuraxial anesthesia may necessitate the use of general anesthesia. If general anesthesia is required, a depolarizing muscle relaxant such as succinylcholine (suxamethonium) should not be given during the period of denervation injury. By a conservative definition, this period begins 24 hours after the injury and lasts for 1 year. The use of succinylcholine during this period of denervation injury may cause severe hyperkalemia. Therefore, a nondepolarizing muscle relaxant should be used to facilitate laryngoscopy and tracheal intubation.

Anticholinesterase drugs, which inhibit the breakdown of acetylcholine, are the mainstay of therapy for all subtypes of the disease. Pyridostigmine , a quaternary ammonium compound that does not cross the blood-brain barrier, is the preferred maintenance drug because it has less severe muscarinic side effects than other anticholinesterase drugs. The treatment goal is full or nearly full remission; most patients will also require immunosuppressive therapy to meet this goal. Prednisone or prednisolone combined with azathioprine are the usual first-line immunosuppressants. Mycophenolate mofetil may also be added; other drugs include rituximab , methotrexate , cyclosporine , tacrolimus , and cyclophosphamide . Monotherapy with glucocorticoids may prevent progression of ocular myasthenia gravis to generalized disease. Thymectomy is performed in all patients with thymomas; however, there is also a benefit of total thymectomy for patients with early-onset symptoms without thymoma.

Myasthenia gravis can manifest in two types of crises. A cholinergic crisis is rare, especially if the daily dose of pyridostigmine is less than 960 mg. It results from an excess of the muscarinic effects of anticholinesterase medications combined with a poor response to anticholinesterase therapy. In contrast, a myasthenic crisis results from a worsening of the disease; its symptoms include worsening muscle weakness, especially muscles of respiration. Admission to an intensive care unit, ventilatory support, and rapid therapies such as plasma exchange and intravenous immune globulin are indicated.

Many drugs can cause a worsening of myasthenic symptoms. These patients are extremely sensitive to drugs that potentiate muscle weakness. These agents include neuromuscular blocking agents , quinidine , propranolol , aminoglycoside antibiotics , and tocolytic agents such as magnesium sulfate and terbutaline . One case report noted worsened symptoms after the maternal administration of betamethasone .

The course of myasthenia gravis during pregnancy varies. In general, approximately 30% of cases improve, 30% worsen, and 30% show no change. The baseline clinical state does not predict the course of the disease during pregnancy, and the course during one pregnancy does not predict the course in a subsequent pregnancy. Because disease exacerbations occur more frequently in the first year after diagnosis of myasthenia gravis, women should be counseled to delay pregnancy for 1 to 2 years after the initial disease diagnosis. There is inconsistent evidence that myasthenia gravis has adverse effects on pregnancy. One case series of 37 pregnancies with myasthenia gravis reported rates of gestational diabetes (30%), premature rupture of membranes (19%), and transient neonatal myasthenia and hyperbilirubinemia (24%), although the absence of a control group means that the specificity of risks for these conditions due to myasthenia cannot be surmised.

A systematic review of 824 pregnancies with myasthenia gravis reported a 56.3% rate of vaginal delivery, an overall exacerbation risk of 33.8%, a 6.4% risk of myasthenic crisis during pregnancy, and an 8.2% risk of myasthenic crisis postpartum. Complications of pregnancy and their treatment (e.g., preeclampsia) may contribute to worsening myasthenia gravis symptoms. For example, magnesium therapy has been reported to precipitate a myasthenic crisis.

The maternal physiologic changes of pregnancy, including alterations in drug absorption, increases in blood volume, and changes in renal clearance, may require adjustments in the doses of anticholinesterase drugs. Each patient should be monitored carefully for progressive respiratory compromise secondary to diaphragmatic elevation during pregnancy. Vital capacity can be measured to monitor fatigue during labor. The treatment of the myasthenic patient with preeclampsia or preterm labor is problematic because the use of magnesium sulfate for maternal seizure prophylaxis or fetal neuroprotection reduces acetylcholine release and may be associated with a significant increase in maternal and fetal muscle weakness, and even maternal death. In those cases, seizure prophylaxis with anticonvulsants has been described.

The uterus consists of smooth muscle; therefore, myasthenia gravis should not affect the first stage of labor. However, the second stage of labor often requires the use of striated muscle, and consequently an assisted (e.g., vacuum or forceps) vaginal delivery may be required.

Maternal IgG antibodies to the acetylcholine receptor are transferred across the placenta. Transient neonatal myasthenia gravis occurs in approximately 10% to 20% of infants of mothers with myasthenia gravis. Notably, the risk and severity of neonatal myasthenia are not related to the severity of maternal disease; therefore, all neonates of mothers with myasthenia gravis must be closely monitored. The symptoms, including poor sucking, generalized hypotonia, difficulty feeding, feeble cry, ptosis, and respiratory distress, usually develop within the first 12 to 48 hours after birth and abate as the antibodies are metabolized, with resolution typically occurring within 2 to 4 weeks. Arthrogryposis multiplex congenita is a rare and severe disease, occurring in about 6.2 per 100,000 live births. In this condition, infants have multiple joint contractures from reduced movement in utero ; the condition is associated with stillbirth, spontaneous abortion, and neonatal death.

Myasthenia gravis patients should undergo early antepartum consultation with an anesthesia provider. This evaluation should assess the extent of bulbar and respiratory involvement and overall baseline muscle strength. Pulmonary function testing should be performed in patients with evidence of respiratory compromise. In a study of surgical patients, the presence of bulbar symptoms, a preoperative serum level of antiacetylcholine receptor antibody > 100 nmol/L, and intraoperative blood loss > 1000 mL were risk factors for having a postoperative myasthenic crisis.

Patients with respiratory compromise may be more susceptible to opioid-induced respiratory depression, and consideration should be given to minimizing or avoiding opioids when possible. Neuraxial techniques are the preferred method for labor analgesia in patients with myasthenia gravis, and early labor neuraxial analgesia is recommended to attenuate stress and preserve maternal strength and reserve for the second stage of labor. The use of anticholinesterase drugs may prolong the half-life of ester local anesthetic agents.

Neuraxial anesthetic techniques are preferred for cesarean delivery unless the patient has significant bulbar involvement or respiratory compromise. The use of bilevel positive airway pressure for ventilatory support in patients with moderate respiratory compromise may improve the safety of neuraxial anesthesia.

In the patient with severe bulbar involvement or respiratory compromise, it may be prudent to secure the airway before surgery. Depolarizing muscle relaxants (e.g., succinylcholine ) have an unpredictable effect in these patients, with affected and unaffected muscles being more sensitive and resistant to these agents, respectively. However, the commonly administered dose of succinylcholine (1 to 1.5 mg/kg), which is three to five times the dose required to depress neuromuscular function by 95%, will most likely provide adequate relaxation even for resistant muscles. Anticholinesterase agents and plasmapheresis cause decreases in the activity of plasma cholinesterase and may cause delays in succinylcholine hydrolysis.

Myasthenic patients are extremely sensitive to nondepolarizing muscle relaxants. If a nondepolarizing muscle relaxant must be given, the anesthesia provider should administer a small amount of an agent with a short half-life (e.g., rocuronium , atracurium , vecuronium ). Mivacurium is metabolized via plasma pseudocholinesterase, which may be inhibited by pyridostigmine. In general, greater disease severity corresponds with enhanced sensitivity to nondepolarizing muscle relaxants, necessitating the use of clinical judgment and neuromuscular monitoring to determine the amount and timing of drug doses. Myasthenia may prevent a full-strength contraction with nerve stimulation; therefore, a control train-of-four stimulus test should be performed before paralysis for later comparison. For nondepolarizing agents, approximately 50% of the standard dose may be adequate, and a prolonged recovery should be anticipated. Sugammadex , a cyclodextrin that works by encapsulating nondepolarizing neuromuscular blocking agents, may be preferred for neuromuscular blockade reversal in patients with myasthenia gravis. The limited available data on its use in pregnancy suggest that it is safe.

After delivery, fluid shifts and decreased maternal alpha-fetoprotein concentration may necessitate an adjustment of anticholinesterase drug doses. Some patients who receive general anesthesia require postoperative ventilation. The following factors are predictive of an increased risk for postoperative ventilation: (1) female sex; (2) FEF _{25% to 75%} (forced expiratory flow during the middle half of the forced vital capacity) < 3.3 L/s and less than 85% of that predicted; (3) FVC (forced vital capacity) < 2.6 L/s and less than 78% of that predicted; and (4) MEF _50% (maximal expiratory flow at 50% of expired vital capacity) < 3.9 L/s and less than 80% of that predicted.

A variety of antiseizure medications are used for seizure therapy, depending on the type of seizure and clinical response. A variety of adverse effects have been reported with these medications, including early-onset events (e.g., somnolence, dizziness, hypersensitivity, rash, gastrointestinal symptoms) and late-onset events (e.g., depression, leukopenia, aplastic anemia, thrombocytopenia, megaloblastic anemia, hyponatremia).

Prognosis for medical control of seizures is good for patients with generalized seizure disorders; as many as 20% to 40% of newly diagnosed epilepsy patients become seizure-free without or with minimal antiseizure medication therapy. Two in three newly treated epilepsy patients will eventually enter long-term remission (≥5 years without a seizure). However, about one-third of patients will have an intermittent pattern (early remission with late recurrence or late remission). Finally, the observed number of deaths in the study compared with the general population is consistently increased in the first several years after the diagnosis of epilepsy.

Epilepsy is common in pregnancy, affecting about 1 in 200 parturients. Maternal mortality is greater in women with epilepsy. Pregnancy may increase the frequency of seizures in one in three epileptic patients; this may be secondary to increases in plasma volume and higher drug clearance, which reduces serum levels of antiepileptic drugs. Women who are free of seizures for at least 9 months to 1 year before pregnancy have a probability of 84% to 92% of remaining seizure-free during pregnancy.

Optimizing antiseizure medication therapy before and during pregnancy is critical to maintain prepregnancy serum antiseizure medication levels because anticonvulsant drug levels can decrease during pregnancy, owing to decreased plasma protein binding and greater drug clearance. Because of the teratogenic potential of many antiseizure medications, some physicians consider withdrawal of these drugs after 2 years without seizures and recommend waiting at least 6 additional months after withdrawal before attempting to conceive. Substituting new antiseizure medications after conception is not recommended because of increased risk for teratogenicity associated with polytherapy.

A variety of causes have been proposed for the increase in seizure frequency observed in some pregnant women ( Table 47.2 ). Higher estrogen concentrations in pregnancy lower the seizure threshold. Greater sodium and water retention, alkalosis secondary to hyperventilation, sleep deprivation, and increased stress and anxiety also have been suggested as mechanisms. The American Academy of Neurology and American Epilepsy Society concluded that monitoring of lamotrigine , carbamazepine , and phenytoin levels should be considered during pregnancy. Monitoring of levetiracetam and oxcarbazepine (and its active metabolite, monohydroxy derivative) may be considered, and monitoring of other antiepileptic agents should not be discouraged despite limited data regarding their pharmacokinetic behavior during pregnancy.

Seizures in a pregnant patient can have devastating consequences. Hypoxia and acidosis that occur during a generalized seizure can result in fetal compromise or intrauterine fetal death. Placental abruption may occur. During the past three decades, the overall risk for obstetric complications in epileptic women has declined. Although some studies have suggested an increased risk for hypertension in pregnancy (including preeclampsia), bleeding, and preterm birth, a review by a panel of experts concluded that evidence is inconclusive. However, women with epilepsy may have a moderately increased risk for cesarean delivery.

Fetuses and neonates of women with epilepsy are more likely to have adverse pregnancy outcomes, including preterm birth, small for gestational age infant, and requirement for labor induction or cesarean delivery. Data are insufficient to judge whether intrauterine exposure to antiseizure medications, versus epilepsy itself, increases the risk for cognitive impairment in the offspring of women with epilepsy, although there is some evidence that the risk may be increased for specific drugs. Specifically, intrauterine exposure to valproate has been associated with maladaptive childhood behavior and autism.

The risk for congenital malformations in women with epilepsy receiving antiseizure medication monotherapy is estimated to be 4% to 6%. Malformations have been previously associated with antiseizure medication (e.g., cleft lip and palate, and cardiac, neural tube, and urogenital defects, specifically hypospadias). A 2016 systematic review and metaanalysis of 31 studies of antiseizure medication therapies in pregnant women concluded that lamotrigine and levetiracetam carried the lowest overall risk for malformation, although data on specific malformations were insufficient. Several studies have suggested that maternal folic acid supplementation before conception may decrease the risk for major congenital abnormalities in the offspring of women with epilepsy who are receiving antiseizure medications.

In 2018 Tomson and Battino published a review of data from the International Registry of Antiepileptic Drugs and Pregnancy and concluded that the risks of major congenital malformation associated with lamotrigine, levetiracetam, and oxcarbazepine were not different from offspring unexposed to antiepileptic drugs.

Neonates of mothers undergoing long-term antiepileptic therapy may be at risk for deficiencies in vitamin K–dependent clotting factors or other coagulation defects, despite the absence of clinically evident maternal coagulation abnormalities. Enzyme-inducing antiseizure medications (e.g., phenytoin , phenobarbital , carbamazepine ) can cross the placenta and may increase the rate of oxidative degradation of vitamin K in the fetus. Affected infants are at risk for neonatal hemorrhage and respond to typical vitamin K doses (1 mg) given intramuscularly at birth. The administration of prenatal vitamin K to epileptic women with long-term exposure to these antiseizure medications has not been conclusively shown to reduce the risk for neonatal hemorrhage.

There are significant interactions between antiseizure medications and anesthetic agents. Carbamazepine , phenytoin , phenobarbital , and primidone are potent inducers of the cytochrome P450 enzymes in hepatic metabolism, which may result in decreased plasma concentrations of many medications, including neuromuscular blockers, beta-adrenergic receptor antagonists, and calcium entry–blocking agents.

Serum levels of antiseizure medications should be checked if therapeutic levels are known. If the patient experiences a seizure during labor, airway protection and support of ventilation are essential. Small doses of a benzodiazepine or propofol arrest most seizures. Fetal bradycardia, typically self-limited, may necessitate immediate delivery.

Oral antiseizure medications should be continued whenever possible. Unfortunately, many of the agents are not available in parenteral forms. If oral agents cannot be taken, conversion to a parenteral agent such as phenytoin may be required. In general, antiseizure medications have sedating properties.

The presence of epilepsy is not a contraindication to the administration of neuraxial analgesia or anesthesia. Although antiseizure medications have been associated with adverse effects on the coagulation system, Manohar et al. observed no abnormal clotting parameters or platelet counts preoperatively in a series of patients with epilepsy undergoing surgery.

If general anesthesia is necessary, it seems prudent to avoid meperidine , as this medication has been associated with seizures even after low to moderate doses. Sevoflurane has stronger epileptogenic properties than isoflurane , but limiting the concentration to less than 1.5 minimum alveolar concentration minimizes risk, and coadministration of nitrous oxide and hyperventilation both counteract this effect. The highest incidence of seizure activity with induction of anesthesia is believed to occur with etomidate . Ketamine may also facilitate seizures at low doses, but at high doses each of these induction agents acts as an anticonvulsant.

In pregnant patients with myotonic dystrophy, symptoms of weakness and myotonia usually remain unchanged; however, in a minority of these women, symptoms worsen during pregnancy but generally resolve after delivery. Antepartum evaluation should include pulmonary function testing to assess the severity of restrictive lung disease caused by muscle wasting and an electrocardiogram, which may reveal conduction abnormalities. Patients with myotonic dystrophy type 1 should be monitored for malignant cardiac arrhythmias and peripartum cardiomyopathy.

There may be a higher risk for preterm labor in patients with myotonic dystrophy. Other complications of pregnancy include polyhydramnios (secondary to reduced fetal swallowing) and an increased risk for placenta previa and accreta. Magnesium sulfate has been reported to cause maternal respiratory compromise. Poor uterine contractions may result in prolonged labor, uterine atony, and an increased risk for postpartum hemorrhage. Skeletal muscle weakness increases the risk for prolonged second stage of labor and operative delivery. The neonate also may have respiratory distress if affected by congenital myotonic dystrophy.

There are reports of patients with myotonia congenita who experience temporary worsening of symptoms during pregnancy. Obstetric problems have not been described, most likely because this disease involves skeletal muscle only; uterine smooth muscle is not affected in these patients.

Many reports of anesthetic management of patients with myotonic dystrophies were published before the distinction between DM1 and DM2 was appreciated, and before the era of modern anesthesia management; therefore, generalization from these reports and application to individual patients may not be valid. The prolonged contractions witnessed in patients with myotonia are caused by an intrinsic muscle disorder that is not relieved by spinal or epidural anesthesia. However, infiltration with a local anesthetic may partially release contractions. Reports suggest that patients with myotonic disorders may be especially sensitive to the respiratory depressant effects of opioids and general anesthetics. Sedative-hypnotic agents should be used with caution as opioids or sedatives may precipitate apnea. Neuraxial anesthesia is preferred for labor and vaginal or cesarean delivery. Both spinal and epidural anesthesia have a long history of successful use in patients with myotonic dystrophy. Although the clinical characteristics of myotonic dystrophy DM2 are generally more benign than DM1, anesthesia providers should be aware that both may be associated with dysphagia, cardiomyopathy, and cardiac conduction abnormalities. Two case reports described atrial flutter and fibrillation resulting in shock after tracheal intubation in patients with myotonic dystrophy receiving general anesthesia for cesarean delivery.

Patients with myotonic dystrophy have a high incidence of pulmonary complications after general anesthesia. Cold external temperatures and shivering are known triggers of myotonia, so the patient should be kept warm.

If general anesthesia is required, it may be prudent to limit the use of opioids and carefully titrate muscle relaxants to mitigate the risk for postoperative pulmonary complications. Depolarizing agents such as succinylcholine should be avoided because fasciculations trigger myotonia, thereby making ventilation and tracheal intubation difficult. In contrast, patients with myotonic dystrophy appear to have a normal response to nondepolarizing muscle relaxants . Regardless, careful neuromuscular monitoring is essential, particularly in those with significant baseline muscle weakness. Reversal of neuromuscular blockade with sugammadex has been described. In a review of dystrophic myotonias and their possible association with malignant hyperthermia, Parness et al. concluded that susceptibility to malignant hyperthermia in this group of patients is similar to that of the general population. Patients with central core disease should be assumed to be susceptible to malignant hyperthermia.

The presentations of dystrophinopathies are variable and the overall management is guided by the presence and severity of symptoms. If significant weakness is present, pulmonary function testing should be obtained to assess the extent of restrictive disease. An antepartum electrocardiogram and echocardiogram should be considered. Pregnant women with muscular dystrophies may have an increased incidence of operative delivery; the presence of severe pelvic wasting may necessitate instrumental vaginal or cesarean delivery. A study of the course of pregnancy in women with hereditary neuromuscular disorders reported a high rate (27%) of abnormal fetal presentation in women with limb-girdle muscular dystrophy, especially in chair-bound patients. Muscle weakness worsened during and after pregnancy in one-half of patients with limb-girdle dystrophy.

Limb-girdle muscular dystrophy is associated with various cardiac abnormalities, including cardiomyopathies and conduction abnormalities. Reduced lung function and respiratory compromise can be exacerbated by the physiologic changes of pregnancy; the requirement for noninvasive positive-pressure ventilation for progression of severe restrictive pulmonary disease during the third trimester of pregnancy in a parturient with limb-girdle muscular dystrophy has been reported. Neuraxial techniques are preferred for labor analgesia and cesarean delivery anesthesia. Severe disease may result in both airway abnormalities and spinal deformities, which may complicate the administration of either general or neuraxial anesthesia.

Although most female carriers of the abnormal gene for muscular dystrophies are asymptomatic, approximately 2.5% have mild symptoms of the disease. Cardiac manifestations may be significant. One report described a female carrier of Becker’s muscular dystrophy who had peripartum cardiomyopathy; general anesthesia was associated with intraoperative respiratory insufficiency, hemodynamic instability, and cardiac arrest. There are reported cases of muscular dystrophy associated with “malignant hyperthermia–like” reactions. In a systematic review, Gurnaney et al. summarized reported cases of patients with muscular dystrophy who developed hyperthermia, tachycardia, rhabdomyolysis, and hyperkalemia after exposure to succinylcholine and/or volatile anesthetic agents . However, none of these patients had other classic signs of malignant hyperthermia or evidence of hypermetabolism. The mechanism for this response may be related to the ability of these agents to exacerbate instability and permeability of dystrophin-deficient muscle membranes. Although muscular dystrophy does not increase risk for malignant hyperthermia , volatile anesthetic agents should be used cautiously because of the risk for severe rhabdomyolysis. Succinylcholine may lead to life-threatening hyperkalemia because of upregulation of extrajunctional acetylcholine receptors or because of rhabdomyolysis and should not be administered to patients with known muscular dystrophy. In general, these patients have a normal response to nondepolarizing muscle relaxants , but careful neuromuscular monitoring is needed, especially in patients with severe muscle wasting.

Neurofibromatosis occurs because of excessive proliferation of neural crest elements such as Schwann cells, melanocytes, and fibroblasts. Clinical manifestations include hyperpigmented lesions ( café-au-lait spots) accompanied by a variety of cutaneous and subcutaneous tumors. This disorder is now believed to exist in two distinct forms with gene abnormalities on two different chromosomes. Neurofibromatosis type 1, the “classic” form, has an incidence of approximately 1 in 3000. The severity and progression of the disease are variable, with the neurologic symptoms depending on the location of the tumors. Intracranial tumors and paraspinal neurofibromas are a cause for concern and may require surgical excision. The risk for pheochromocytoma is greater in these patients. Neurofibromatosis type 2 is a less common form of the disease with fewer cutaneous lesions. Acoustic neuromas as well as other cranial or spinal neurofibromas, meningiomas, and gliomas may be present.

Tuberous sclerosis is characterized by epilepsy, mental disability, and adenoma sebaceum. The brain shows abnormal growth of glial cells in hamartomas called tubers. Hamartomatous tumors can occur in multiple organs, including the heart, kidneys, liver, and lungs. The inheritance pattern is autosomal dominant with a variable expression, and the disease is slowly progressive.

One group of phakomatoses consists of disorders in which a cutaneous vascular anomaly is accompanied by CNS abnormalities (see Box 47.2 ). There are few reports of pregnancy in patients with these disorders. Patients may have neurologic impairment related to hemangiomas of the spinal and cranial CNS. Epidural and spinal anesthesia have been reported in only a few patients with spinal hemangiomas; epidural anesthesia may be preferable to spinal anesthesia as von Hippel-Lindau hemangioblastomas are not present in the epidural space. The presence of widespread varicosities in patients with these disorders may result in a chronically low ventricular preload; if a significant increase in preload occurs during the peripartum period, cardiac overload and peripartum cardiomyopathy may occur.

The incidence of this syndrome appears unchanged in pregnant women, but increases in the first 3 months postpartum. There is no evidence that pregnancy affects the natural history of the disease, and termination of pregnancy does not appear to improve its course.

Anesthetic management depends on patient status at the time of delivery; epidural and spinal techniques have been described in patients with Guillain-Barré syndrome. However, some anesthesia providers have expressed concern regarding the use of neuraxial techniques in these patients, citing the theoretical potential for adverse neurologic effects as a result of anesthetic toxicity or immunologic modulation. Steiner et al. implicated epidural anesthesia as a trigger of Guillain-Barré syndrome in four patients. Wiertlewski et al. reported the immediate worsening of neurologic status after delivery in a pregnant patient with Guillain-Barré syndrome who had received epidural anesthesia. These authors did not establish a causal relationship between the disease and neuraxial anesthetic techniques, nor did they properly acknowledge the increased frequency of Guillain-Barré syndrome in the postpartum period.

If general anesthesia is necessary, succinylcholine should be avoided because of the risk for hyperkalemia in patients with acute muscle wasting. Careful titration of nondepolarizing muscle relaxants is also necessary.

The parturient with a history of remote Guillain-Barré syndrome may have persistent diminished respiratory reserve, even in the absence of obvious residual disability. Pulmonary evaluation should be considered before the administration of anesthesia. Approximately 5% of patients experience a relapse, with a small number of cases progressing to a chronic disorder.

Although the poliovirus vaccine has been available since the 1950s, the last phase of poliomyelitis eradication has been difficult; in 2014 the WHO declared the worldwide transmission of polio virus to formerly polio-free countries to be a public health emergency. The oral form of the vaccination does not appear to have harmful effects on fetal development and can be used if vaccination is required during pregnancy. A history of poliomyelitis may increase the risks for preeclampsia, maternal renal dysfunction, low-birth-weight infants, perinatal death, and cesarean delivery in poliomyelitis survivors. Some of these adverse outcomes may be related to chronic pulmonary issues or mechanical obstruction during labor.