Long considered the safest drug in pregnancy, recent studies have raised concerns that the use of acetaminophen during pregnancy is linked to neurobehavioral disorders such as attention deficit hyperactivity disorder (ADHD) and autism spectrum disorder (ASD) in children. The studies suggesting this association often have serious limitations, including relying on self-reported data and inadequate control for confounding factors. A rigorous analysis of all singleton liveborn children born in Sweden between 1995 and 2019 (n = 2,489,721), which controlled for a large number of confounders and included a sibling analysis, did not find a link between acetaminophen use during pregnancy and offspring risk for ADHD, ASD, or intellectual disability.

Despite a general recommendation to avoid the use of NSAIDs during pregnancy unless the maternal benefit is deemed to significantly outweigh the risk to the fetus, over-the-counter NSAID use is common during pregnancy. In one multicenter survey of more than 5000 pregnant women, 24% reported the use of ibuprofen, 5% reported the use of aspirin, and 4% reported the use of naproxen.

Current data suggest no effect of NSAIDs use on early pregnancy loss. NSAIDs should be used sparingly and avoided if possible in the first trimester. Large observational studies have suggested a small increase in the risk for congenital defects, generally small cardiac septal defects, neural tube defects, and possibly gastroschisis. Compared with neonates exposed to acetaminophen alone, neonates exposed to NSAIDs in the first trimester were found to have an increased risk for gastroschisis, hypospadias, cleft lip and palate, anencephaly, spina bifida, hypoplastic left heart syndrome, pulmonary valve stenosis, and tetralogy of Fallot. An association does not demonstrate causation, and it is possible that the increased risk for congenital anomalies may be related to the indication for use rather than the pharmacologic effect of NSAIDs. The US Food and Drug Administration (FDA) warns against the use of NSAIDs at 20 weeks’ gestation or later because they can cause fetal renal dysfunction and low amniotic fluid volume, in addition to ductus arteriosus constriction. Unless the maternal benefits significantly outweigh the risks, NSAIDs are generally contraindicated in the third trimester because of increased risk for premature closure of the ductus arteriosus as well as fetal renal dysfunction and oligohydramnios.

Historically, the ACOG warned against the postpartum use of NSAIDs in women recovering from preeclampsia with persistent postpartum hypertension; however, following a systematic review and metaanalysis showing that postpartum NSAID use among women with hypertensive disorders of pregnancy was not associated with maternal hypertension exacerbation, this recommendation has been updated and the ACOG no longer discourages NSAID use in these women.

NSAIDS are an important component in the multimodal management of postpartum pain in patients with chronic pain and opioid tolerance. They are widely utilized and extremely effective for postdelivery pain. Given the limited transfer of ibuprofen to breast milk, the Academy of Breastfeeding Medicine deems it a suitable analgesic drug in breastfeeding mothers. Data on other NSAIDs (e.g., etodolac, fenoprofen, meloxicam, oxaprozin, piroxicam, sulindac, and tolmetin) are limited, and the FDA labeling discourages their use. Less than 1% of the maternal dose of ketorolac is excreted into human milk, and adverse events in nursing mothers have not been reported.

Low-dose aspirin is increasingly used beginning at 12 weeks’ gestation until delivery for preeclampsia prophylaxis in women at high risk for the disease (see Chapter 35 ). To date, no significant adverse fetal effects have been noted in the offspring of women allocated to aspirin compared with placebo therapy in large clinical trials.

Cyclooxygenase-2 (COX-2) inhibitors are additive or synergistic with opioids and other analgesics. Compared with nonselective NSAIDs, COX-2 inhibitors have minimal effects on platelet adhesion and are less likely to be associated with bleeding and coagulopathy. However, concerns about potential increased risk for cardiovascular and thrombotic events, combined with the baseline elevated risk for these events during pregnancy and the postpartum period, have prevented COX-2 inhibitors from playing a major role in postpartum analgesia. In addition, the use of celecoxib in late pregnancy has been associated with an increased risk for preterm delivery (adjusted odds ratio [aOR], 2.46; 95% confidence interval [CI], 1.28 to 4.72). In 2023, the Society for Maternal-Fetal Medicine consult series on systemic lupus erythematosus recommended that COX-2 inhibitors, as well as high-dose aspirin, be avoided during pregnancy.

Gabapentin is a mainstay in the management of chronic pain. It is thought to provide analgesia by binding to presynaptic voltage-gated calcium channels in the dorsal root ganglia of the spinal cord, thereby inhibiting the release of excitatory neurotransmitters. Perioperative gabapentin has been shown to decrease acute pain after a variety of surgical procedures, although recent data do not support its routine use to reduce acute or chronic postoperative pain in adult surgical patients. Similarly, data from healthy women undergoing elective cesarean delivery are conflicting.

Gabapentin crosses the placenta. Studies on its use during pregnancy show a wide range of fetal-to-maternal (F/M) drug ratios, likely reflecting the timing of sampling after delivery. The Gabapentin Pregnancy Registry examined the outcomes of 39 pregnancies in which mothers received gabapentin for epilepsy or chronic pain conditions. Thirty-six of these mothers received gabapentin continuously throughout pregnancy; there was no evidence of increased risk for adverse maternal or fetal/neonatal effects compared with the general population. The most recent population-based study evaluating the association between gabapentin exposure during pregnancy and risk of adverse neonatal and maternal outcomes used the US Medicaid Analytic eXtract (MAX) data set. Gabapentin exposure during early pregnancy was not associated with major malformations, although a higher risk of cardiac malformations was identified. Late pregnancy gabapentin exposure was associated with preterm birth, small-for-gestational-age infants, and neonatal intensive care unit admission.

The milk/plasma (M/P) ratio of gabapentin was assessed in five lactating mothers whose chronic gabapentin dose ranged from 600 to 2100 mg/day; the mean ratio was 1.0 (sampling ranged from 2 weeks to 3 months of neonatal age). The infant dose (mg/kg body weight) of gabapentin was estimated to be 1.3% to 3.8% of the maternal dose. The drug is rapidly metabolized by the newborn ; mean neonatal plasma concentrations were 27% of the umbilical artery plasma levels at 24 hours postpartum. No adverse effects were observed in the five mother-infant dyads. Thus, current data suggest that gabapentin use by lactating mothers is safe.

Pregabalin is not as well studied as gabapentin for the treatment of chronic pain in pregnancy. Although concerns were raised for a possible link between pregabalin use in the first trimester and increased risk for major fetal malformations, follow-up studies have proven more reassuring. A 2016 study included 116 pregnancies with first-trimester exposure; 6% (n = 7) of offspring had malformations compared with 2% (n = 2) in the nonexposed group. Four of the six fetal malformations in the pregabalin group were cerebral ventriculomegaly. By contrast, results from a subsequent larger cohort study from the MAX database (477 infants exposed to pregabalin in the first trimester) were reassuring. After propensity score adjustment for indication, the relative risk (RR) for major fetal malformations in the pregabalin group was 1.16 (95% CI, 0.81 to 1.67). When limited to patients on pregabalin monotherapy, the relative risk for malformations was 1.0 (95% CI, 0.7 to 1.5). Additionally, none of the 477 pregabalin-exposed infants carried a diagnosis of cerebral ventriculomegaly or other brain anomalies reported in the initial smaller study.

Pregabalin crosses into breast milk, as predicted by its low molecular weight and lack of plasma protein binding. In 10 lactating women receiving daily pregabalin 300 mg, the calculated amount of pregabalin present in 24-hour breast milk collections was 0.2% of the administered daily maternal dose, and no adverse effects were reported.

Antidepressants are commonly used to treat chronic pain conditions ranging from migraine disorders to fibromyalgia and neuropathic pain conditions. Two classes of antidepressants have demonstrated efficacy in the management of chronic pain: tricyclic antidepressants (TCAs) and serotonin-norepinephrine reuptake inhibitors (SNRIs). Both increase the synaptic concentrations of serotonin and norepinephrine.

Among TCAs, amitriptyline and nortriptyline have the largest body of reassuring safety data in pregnancy and lactation. TCAs cross the placenta; the approximate F/M ratio is 0.6. The largest and most recent epidemiologic studies on the risk for major congenital malformation have yielded conflicting results. Among the TCAs, clomipramine has been the most strongly implicated in increased risk for congenital anomalies and postexposure withdrawal syndrome in neonates.

Duloxetine and venlafaxine are commonly used SNRIs for chronic pain conditions. In a recent Scandinavian study, the chronic use of SNRIs during pregnancy was not associated with adverse pregnancy outcomes, and duloxetine exposure was unlikely to increase the risk of small-for-gestational age. Like the TCAs, neonates of mothers taking SNRIs at the end of pregnancy should be monitored for transient agitation and withdrawal symptoms; the risk is greater in drugs with shorter half-lives, such as venlafaxine. The American Academy of Pediatrics (AAP) recommends using SNRIs with caution in lactating mothers because the long-term effects of chronic exposure on neurodevelopment are unknown.

Ketamine , an N -methyl- d -aspartate (NMDA) receptor antagonist, is used for the treatment of acute postoperative pain and prevention or reversal of central sensitization in the setting of chronic pain.

Ketamine is contraindicated in pregnancy, given lack of safety data and its potential effects on the developing fetus. Animal studies suggest that prenatal and early postnatal exposure to ketamine can be neurotoxic to the developing brain, with long-term neurocognitive abnormalities in offspring of pregnant rats treated with ketamine. Similar effects on the developing fetal brain after exposure to esketamine have recently been suggested.

Ketamine is sometimes used for intraoperative analgesia as an adjuvant to neuraxial anesthesia or for induction of general anesthesia in cesarean deliveries (see Chapter 26 ); small intravenous doses provide significant analgesia with minimal respiratory depression. It is reasonable to consider using ketamine in a patient with chronic pain or to prevent conversion to chronic pain after cesarean delivery. In one study, patients undergoing cesarean delivery with spinal anesthesia and multimodal postoperative analgesia that included spinal morphine were randomized to receive ketamine 10 mg or placebo after delivery. Women who received ketamine reported less pain at 2 weeks postpartum compared with women who received a placebo.

There are currently limited data regarding ketamine safety in lactation. Two studies on the pharmacokinetics of ketamine transfer in breast milk suggest that transfer of ketamine and its major metabolites (norketamine, dehydronorketamine, and hydroxynorketamine isomers) into human milk is minimal, with an estimated relative infant dose (RID) of less than 1% for ketamine and norketamine.

The use of neuraxial clonidine and, more recently, dexmedetomidine , as adjuvants for intraoperative and postcesarean analgesia has been reported. In a 2024 statement from the American Society of Anesthesiologists, consideration for the use of epidural clonidine and dexmedetomidine as adjuvants for the management of intraoperative pain during cesarean delivery with epidural anesthesia is suggested. The epidural formulation of clonidine carries a black box warning from the US FDA stating the following: “Clonidine hydrochloride injection (epidural clonidine) is not recommended for obstetrical, postpartum or perioperative pain management. The risk of hemodynamic instability, especially hypotension and bradycardia, from epidural clonidine may be unacceptable in these patients. However, in a rare, obstetrical, postpartum or perioperative patient, potential benefits may outweigh the possible risks.” The use of intrathecal clonidine is off-label.

Neuraxial clonidine may be beneficial in patients who are at risk for the development of neuropathic pain after cesarean delivery. One study reported that the combination of intrathecal clonidine (150 μg) and bupivacaine was associated with decreased wound hyperalgesia after cesarean delivery ; however, further study of the use of alpha ₂ -adrenergic agonists in women with chronic pain is warranted. The use of neuraxial (both epidural and intrathecal) dexmedetomidine is off-label for all patient populations and indications.

Neuromodulation, such as SCS, is a reversible analgesic therapy that is being more widely used because systemic adverse effects are minimal. The basis of neuromodulation was established by the gate theory of pain, highlighting the role of the dorsal spinal cord columns in pain perception. In addition to altering pain perception, SCS also improves microcirculation and activates dorsal column afferent neurons while inhibiting sympathetic efferent neurons, resulting in modification of neurotransmitter activity at the segmental and supraspinal levels. In painful ischemic conditions, SCS is purported to improve oxygenation by vasodilation resulting from reduced sympathetic tone.

SCS is based on stimulation of the dorsal column of the spinal cord through one or more electrodes (SCS leads) permanently implanted in the epidural space. A typical SCS lead has 8 to 16 contacts used to generate a customized electric field. Unilateral or bilateral leads are placed in the dorsal epidural space from C3 to T11 depending on the area and laterality of pain. The lead placement corresponds with the dermatomal distribution of pain: cervical for neck and upper extremity pain and mid- to low-thoracic for low back and lower extremity pain. These leads are anchored with sutures to the supraspinous ligament and tunneled to the implanted pulse generator in a subcutaneous pocket positioned in the patient’s lower back. It is important to understand the number, type, and location of leads prior to considering any neuraxial procedure.

Despite the absence of Class 1 evidence for the effectiveness of neuromodulation in pain management, its use has increased. The FDA has not approved neuromodulation systems during pregnancy, and there is no established consensus on managing patients with such devices during and after pregnancy, and during labor and delivery. Because evidence on the safety of SCS during pregnancy is scarce, out of an abundance of caution, current industry recommendations are to deactivate SCSs throughout pregnancy and in the intraoperative period.

There are two commonly cited concerns with SCS or DRG therapy in pregnancy. The first concern is the impact of the electromagnetic field (EMF) generation around the implanted leads on the developing fetus. Teratogenic effects and reduction in fertility from chronic exposure to EMF is a theoretical concern. Clinically, the magnitude and extent of the EMF are insignificant, and it is highly unlikely that it exerts any substantive ill effects on pregnancy. Most studies suggest that neuromodulation systems produce minimal electrical currents, and the EMF is confined to the posterior spinal column and is insulated from vertebral bones and ligaments. More studies are needed to establish stimulation parameters of these devices and the miscarriage and teratogenic rates in pregnant patients.

A second concern is impairment of placental blood flow, although studies demonstrate that SCS improves microcirculation and produces peripheral vasodilation without compromising uteroplacental blood flow. The uteroplacental circulation lacks autonomic innervation and lacks the regulatory mechanisms of the peripheral circulation. Thus, it seems unlikely that the autonomic effects of SCS will have any perceptible impact on placental or fetal blood flow. Ultrasonographic studies showed no notable differences in uteroplacental and fetal blood flows during the on and off phases of SCS therapy. Human data on the lack of effect of SCS on the developing fetus are limited to a few case reports.

Ultimately, patients should be educated on the potential ramifications of neuromodulation and intrathecal drug therapy. Shared decision-making around care should be made with collaboration and communication among a multidisciplinary team consisting of the anesthesiologist, obstetrician, surgeon, nursing team, pharmacist, pain medicine physician, and the patient (see Chapter 32 ).

A case series indicated that SCS lead breakage associated with pregnancy and childbirth was most likely due to increased mechanical stressors associated with increased abdominal pressure during pregnancy and delivery, and that the abdominal wall should be avoided as a site for the pulse generator implantation in women planning pregnancy. Switching off the SCS or DRG stimulator mitigates accidental reprogramming. Device manufacturers also recommend that patients with implanted SCS should avoid procedures that potentially interfere with device programming such as high-output ultrasound or lithotripsy. Manufacturers recommend (1) turning off the neuromodulation device, (2) avoiding the focus of high-output ultrasound or lithotripsy beam within 15 cm of the SCS, and (3) interrogating the device in the postoperative period before turning it back on.

Turning off the SCS is not an inconsequential decision; poor analgesia may have deleterious consequences in pregnancy. Worsening pain can make pregnancy challenging; patients may have to judiciously resume or increase systemic analgesic intake or conversely, they may choose to endure pain to avoid potential analgesic side effects. Both these choices may have subtle or significant psychosocial and physiologic adverse effects on the mother and fetus.

Lumbar neuromodulation devices in pregnant women create a unique problem for the anesthesia provider. SCS leads in the epidural space can make the administration of neuraxial analgesia and anesthesia challenging. There is a risk of accidental injury to epidural leads with neuraxial procedures, but the presence of neuromodulation devices should not dissuade the anesthesia provider from performing a neuraxial anesthetic. Antepartum referral to the anesthesia provider is optimal as it allows adequate time for review of records and appropriate planning. Review of the manufacturer’s recommendations is essential.

A 2019 systematic review evaluated 18 publications reporting on 25 pregnant patients with 32 pregnancies with SCS; complex regional pain syndrome (CRPS) and failed back were the most common indications. The first report of management of a SCS in a pregnant woman in 1999 was in the setting of CRPS after a traumatic brachial plexus injury. Prior to the pregnancy, she was treated with chronic systemic medications that she wanted to avoid during pregnancy. The patient was offered a cervical SCS. The device was placed before pregnancy and resulted in excellent pain relief and discontinuation of all pain medications. The SCS was not turned off during the peripartum period. Since this initial report, there have been many additional reports of pregnant women with indwelling SCS, including cervical and thoracic or lumbar SCS with descriptions of neuraxial analgesia and anesthesia management for vaginal or cesarean delivery.

In one case series, one of four patients had failed neuraxial anesthesia. The possibility of fibrosis created by the SCS leads forming an encapsulating sheath or scarring in the epidural space may increase technical challenges or suboptimal spread of epidural medication, even if the epidural catheter is correctly sited in the epidural space. Spinal anesthesia is not impacted.

The anesthesia provider should review the device implantation procedure report to confirm the site of lead placement, power source location, and the path of the leads and extension wires. If the report or imaging is unavailable, a single X-ray of the lumbar spine may be considered to identify the details of the implanted device. A single plain X-ray exposes the patient and fetus to a minimal ionizing radiation (see Chapter 18 ).

Another concern during cesarean deliveries and other surgical procedures is complications stemming from the use of diathermy or electrocautery. This may damage the neuromodulator lead insulation, suppress stimulator output, and inadvertently reprogram the neurostimulator. Painful electric shocks have been reported; thus, most manufacturers recommend avoidance of monopolar diathermy when possible. Bipolar diathermy appears to be safe. Most practice guidelines suggest decreasing SCS amplitude to zero followed by powering off the device and using bipolar cautery in short, quick intermittent bursts at the lowest feasible energy levels. The electrical dispersal pad should be placed close to the field and far from the SCS components.

Implantable intrathecal pumps (ITPs) are used for the treatment of severe chronic pain or intractable spasticity; medications such as opioids, clonidine, and baclofen are delivered directly into the cerebrospinal fluid. Rarely, the care team may encounter ITP drug delivery devices in pregnant patients. The advantage of these systems lies in the ability to administer high-potency medications while minimizing systemic side effects; the intrathecal dose is substantially smaller than the equipotent/equianalgesic systemic dose (100 to 1000 times more potent, depending on the medication). An ITP system usually consists of a programmable reservoir placed subcutaneously in the lower abdomen and a tunneled catheter of variable length communicating between the reservoir and intrathecal space. The catheter tip is sited in the low thoracic or high lumbar region. There is little evidence or clinical experience to inform the healthcare team of the safety of these systems in pregnant women. The few case reports describing pregnant patients with ITPs have not described any adverse effects on the fetus or significant plasma levels of the medication in the neonate or in breast milk.

The anesthesia team should be knowledgeable of the drugs delivered through the ITP to recognize and mitigate complications associated with ITP drug overdose or withdrawal. Abrupt discontinuation of intrathecal medication has serious consequences. For example, withdrawal from baclofen may cause hallucinations, seizure, hypertonicity, hypermetabolic emergency, multiorgan system failure, and death. It is essential to ensure the ITP has adequate battery life, the reservoir is filled with appropriate medication(s), and the logistics regarding ITP use through pregnancy and peripartum are well organized.

There are no absolute contraindications to the provision of neuraxial analgesia or anesthesia in the patient with an ITP. Awareness of the entry level of the intrathecal catheter will allow the anesthesia provider to identify a spinal interspace (usually the lowest feasible) for the neuraxial anesthetic procedure. The pump system and the level of the ITP catheter can be identified by reviewing the ITP procedure report, communicating with the implanting physician, and reviewing plain radiographs. Other potential concerns are localized discomfort at the reservoir site with increasing abdominal size in pregnancy. Care should be taken during cesarean delivery to prevent violation of the pump pocket. Meticulous sterile precautions during the performance of neuraxial blockade may help prevent infectious complications.

Pharmacologic treatment and prophylactic medications from a migraine headache are summarized in Table 42.1 .

The use of nonpharmacologic interventions on LBP in pregnancy was evaluated in a 2021 systematic review and metaanalysis which included six randomized trials (693 pregnant participants); exercise, muscle relaxation exercises accompanied by music and transcutaneous electrical nerve stimulation (TENS) were more effective than the other types of interventions (kinesio tape, yoga, and ear acupuncture) in the treatment of pregnancy-related LBP.