Anesthesia for cesarean delivery

INTRODUCTION — 

The plan for anesthesia for cesarean delivery (CD) must consider the comfort and safety of the parturient and the well-being of the fetus and neonate. This topic will discuss the management of regional and general anesthesia for CD. Anesthesia for CD in patients with preeclampsia and neuraxial labor analgesia are reviewed separately.

●(See "Anesthesia for the patient with preeclampsia", section on 'Anesthesia for cesarean delivery'.)

●(See "Neuraxial analgesia for labor and delivery (including instrumental delivery)".)

PREOPERATIVE ASSESSMENT

●History and physical examination – All patients who have anesthesia for CD should have a focused history and physical examination by an anesthesia provider [1], including airway assessment and a review of gestational history. (See "Airway management for the pregnant patient".)

The preoperative assessment for emergency CD should be as comprehensive as time permits.

●Antenatal anesthesia consultation – It is reasonable to schedule an antenatal consultation with an anesthesiologist for patients at risk of complications due to preexisting conditions, even if CD is not planned. Each parturient should be viewed as a potential candidate for operative delivery during labor, possibly on an emergency basis. Some of the indications for antenatal anesthesia consultation are shown in a table (table 1).

●Laboratory testing – Preoperative laboratory testing should be individualized, based on patient factors and institutional norms. Pregnant patients usually have a blood type and antibody screen performed as part of routine prenatal care; repeat preoperative type and screen should be performed for patients at high risk of severe bleeding and for those with red blood cell antibodies. Baseline hemoglobin, platelet count, and coagulation testing may be indicated for specific conditions. (See "Cesarean birth: Preincision planning and patient preparation", section on 'Laboratory testing'.)

PREPARATION FOR ANESTHESIA — 

In addition to routine preparation for anesthesia, issues specific to CD include the following:

●Aspiration prophylaxis – The obstetric patient is at increased risk of aspiration of gastric contents, especially in cases of difficult or failed intubation [2]. (See "Airway management for the pregnant patient", section on 'Fasting and aspiration prophylaxis'.)

•Preoperative fasting – Gastric emptying is not changed by pregnancy [3-5], and preoperative fasting guidelines are the same for elective CD as for other surgical procedures. (See "Preoperative fasting in adults", section on 'Pregnant patients'.)

For urgent CD (eg, nonreassuring fetal status, breech presentation in labor, laboring on a surgical scar) in patients who have not fasted, the risks of aspiration must be weighed against the risks of delay of surgery to allow fasting. In many cases, it will make sense to administer pharmacologic aspiration prophylaxis and proceed to CD without delay.

•Pharmacologic prophylaxis – Practice varies with respect to pharmacologic aspiration prophylaxis for CD. Consistent with guidelines from the American Society of Anesthesiologists, we consider administering nonparticulate antacids, H2 receptor antagonists, and/or metoclopramide [1,6].

Options for pharmacologic prophylaxis are shown in a table (table 2) and are discussed separately. (See "Rapid sequence induction and intubation (RSII) for anesthesia", section on 'Premedication'.)

•Point of care gastric ultrasound – Point of care gastric ultrasound may be an option to assess individual patient aspiration risk, though the role of gastric ultrasound in clinical decision making has not been established [7-9]. For clinicians who are experienced in gastric ultrasound, it may be helpful, especially in patients who may not be candidates for neuraxial procedures or in non-fasted patients with risk factors for difficulty with airway management. It should be noted that gastric ultrasound in pregnant patients may vary slightly from non-pregnant patients. (See "Preoperative fasting in adults", section on 'Methods used to measure stomach emptying or volume' and "Overview of perioperative diagnostic uses of ultrasound", section on 'Gastric ultrasound'.)

●Premedication – We discuss the options for anxiolysis with patients prior to entering the operating room. Sedatives are not usually administered prior to CD, as these medications could result in amnesia for the birth, and they cross the placenta. Most patients require no more than reassurance from the anesthesia clinician. For patients with severe anxiety, we find that a single dose of midazolam (1 to 2 mg intravenous [IV]) is often sufficient and is likely safe. In one randomized trial, administration of midazolam (0.02 mg/kg IV) plus fentanyl (0.1 mcg/kg IV) prior to spinal anesthesia for CD resulted in similar neonatal Apgar scores, pulse oximetry, and neurobehavioral scores when compared with placebo [10]. There was no difference in the mother’s recall of the birth.

For patients who are particularly anxious about neuraxial anesthesia placement, 30 to 50 percent nitrous oxide by patient-controlled mask during placement may be an option.

●Fetal monitoring – The fetal heart rate should be documented prior to CD. For laboring patients, fetal heart rate monitoring should continue after transfer to the operating room, to the extent possible. (See "Cesarean birth: Preincision planning and patient preparation", section on 'Fetal heart rate monitoring'.)

CHOICE OF ANESTHETIC TECHNIQUE — 

Options for anesthesia for CD include neuraxial anesthesia (ie, spinal anesthesia, combined spinal-epidural anesthesia (CSE), and epidural anesthesia) and rarely, general anesthesia.

General versus neuraxial anesthesia — For patients for whom either general or neuraxial anesthesia could be used for CD, we and most clinicians strongly prefer to use neuraxial anesthesia. Neuraxial anesthesia is used for >95 percent of CDs in the United States and Canada [11,12]. There are rare exceptions for which general anesthesia (GA) is more appropriate as detailed below.

Both types of anesthesia are very safe, and anesthesia related risks to the mother and baby are very low with both general and neuraxial anesthesia. The choice of anesthetic technique should be based on maternal and fetal status, comorbidities, expected duration and difficulty of the procedure, the presence (or absence) of an in situ epidural or spinal catheter, and patient preferences.

Many patients prefer neuraxial anesthesia because it allows them to be awake for the birth.

If a working epidural catheter is in place, it is typically used to achieve surgical anesthesia. An overall approach to the choice of anesthetic technique for CD is shown in an algorithm (algorithm 1). (See 'Epidural drugs for CD' below and 'Failed or inadequate neuraxial block' below.)

Maternal and neonatal mortality is not clearly improved with neuraxial anesthesia, however maternal morbidity may be less with neuraxial anesthesia [13].

For urgent or emergency CD, spinal anesthesia may be appropriate, particularly for patients who have increased risks associated with general anesthesia (ie, known difficult airway, recent oral intake of food, or malignant hyperthermia susceptibility). A single-shot spinal anesthetic by an experienced clinician does not take much longer than induction of GA. For most patients who have a well-functioning labor epidural in place, the catheter can be used to convert to surgical anesthesia, even for emergency CD, as discussed below.

Choice of anesthetic technique for CD in patients with preeclampsia is discussed separately. (See "Anesthesia for the patient with preeclampsia", section on 'Choice of anesthetic technique'.)

●Advantages of neuraxial anesthesia for CD

•The mother can be awake for the birth and a partner can be present.

•Neuraxial anesthesia avoids the need for airway instrumentation, which may be more challenging in the parturient. This is discussed separately. (See "Airway management for the pregnant patient", section on 'Incidence and consequences of airway problems'.)

•Neuraxial anesthesia minimizes the use of systemic medication and transfer of medication to the fetus.

•Neuraxial anesthesia allows the use of neuraxial opioids for postoperative analgesia, thereby minimizing the need for systemic opioids. (See "Post-cesarean delivery analgesia".)

•Neuraxial anesthesia is associated with less maternal morbidity (eg, perioperative venous thromboembolism, surgical site infection, blood loss, and severe postpartum depression). (See 'Comparative anesthesia related maternal morbidity' below.)

●Indications for general anesthesia – General anesthesia may be indicated for CD in the following scenarios:

•Emergency CD, with insufficient time to perform neuraxial anesthesia or to achieve a surgical level of anesthesia via labor epidural catheter [14]

•Maternal refusal of, or inability to cooperate with neuraxial anesthesia

•Failed neuraxial technique without time to replace (see 'Failed or inadequate neuraxial block' below)

•Anticipated massive hemorrhage or highly complex surgery (see "Anesthesia for the patient with peripartum hemorrhage", section on 'General concerns')

•Some maternal cardiac comorbidities (see "Anesthesia for labor and delivery in high-risk heart disease: General considerations")

•Relative or absolute contraindications to neuraxial anesthesia (eg, coagulopathy, severe hypovolemia, infection at the site of epidural or spinal placement, some intracranial pathology) (see "Overview of neuraxial anesthesia", section on 'Preoperative evaluation' and "Obstetric and nonobstetric anesthesia for patients with neurologic disorders", section on 'Choice of analgesic or anesthetic technique')

Comparative anesthesia related maternal mortality — Anesthesia-related mortality during CD is very rare and is likely similar whether the patient has GA or neuraxial anesthesia [15]. Anesthesia practice has evolved for both neuraxial and general anesthesia (including airway management) for CD:

●In a 2012 review of Pregnancy Mortality Surveillance System in the United States: For the epoch 1997 to 2002, maternal mortality was greater with GA than with neuraxial anesthesia (6.5 versus 3.8 per million deliveries) [15]. Two-thirds of deaths related to GA were caused by intubation failure or problems on induction, and more than 25 percent of deaths during regional anesthesia during CD were due to high neuraxial block.

●In a 2022 retrospective cohort study of 14,750 patients who had GA for CD from 2004 to 2019 from Multicenter Perioperative Outcomes Group, there was one cardiac arrest (etiology unknown) and there were no deaths [16].

●In the 2023 Confidential Enquiries for maternal death in the United Kingdom, in the epoch for 2019 to 2021 there were over two million births and 241 maternal deaths, of which only one was related to anesthesia (tension pneumothorax of unclear etiology during GA) [17].

Comparative anesthesia related maternal morbidity — Some adverse effects may be less common after neuraxial anesthesia than after GA for CD.

●Blood loss during CD may be lower with neuraxial anesthesia, though the difference is likely clinically insignificant, without a difference in transfusion [18-20].

●Neuraxial anesthesia may be associated with reduced incidence of perioperative venous thromboembolism and surgical site infection compared with GA [13,21].

●Neuraxial anesthesia may be associated with a reduced risk of severe postpartum depression. In a retrospective database study including over 400,000 cesarean deliveries in the State of New York, the use of GA was associated with increased risks of postpartum depression requiring hospitalization, suicidal ideation, and self-inflicted injury [22]. However, conclusions are limited by potential confounding by the circumstances that may require general anesthesia (ie, emergency or urgent CD, fetal distress).

●In the Serious Complications Repository project including data on over 25,000 obstetric anesthetics from 2004 to 2009, there were 85 anesthesia related complications [23]. There were no cardiac arrests related to GA and one case of hypoxemic arrest in a patient with a high neuraxial block. (See 'High neuraxial block' below.)

Neonatal effects — The literature on the differences in neonatal outcomes associated with the choice of anesthetic technique for CD is inconclusive, but any effect is likely small. Apgar scores and umbilical acid base status may be affected by variables unrelated to the choice of anesthetic, including the indication for the CD, vasopressors administered during anesthesia (ie, phenylephrine versus ephedrine), surgical technical issues, and others.

A meta-analysis of 22 randomized or quasi-randomized trials including 1800 patients who underwent CD reported no difference in umbilical artery or vein pH during non-urgent CD between patients with neuraxial versus general anesthesia [18]. Mean Apgar scores at one minute were higher with epidural than with general anesthesia, with no differences in five-minute Apgar score between general and any type of neuraxial anesthesia.

In a subsequent single institution retrospective case control study of patients who had complex CD with incision to delivery time ≥10 minutes, 60 patients who had general anesthesia were matched with 60 patients who had neuraxial anesthesia [24]. Apgar scores at 1 and 5 minutes were higher in neonates whose mothers had neuraxial anesthesia, however arterial and venous cord blood pH, rate of neonatal intensive care unit admission, and length of stay were similar in the two groups.

Neonates whose mothers have GA may require stimulation and respiratory support for the first few minutes after delivery. The neonatal resuscitation team should be notified of all medications administered to the mother during induction of GA, and prior to delivery.

General anesthesia prevents bonding between the mother and neonate immediately after the delivery and may delay initial breastfeeding.

NEURAXIAL ANESTHESIA — 

A sensory level from sacral dermatomes to T4 is desired for CD.

Choice of neuraxial technique — For patients without a labor epidural catheter in place and for whom neuraxial anesthesia is appropriate, we suggest using a combined spinal epidural (CSE) or spinal anesthesia for cesarean delivery rather than epidural anesthesia alone. Spinal anesthesia is usually performed more quickly than epidural anesthesia and provides rapid onset of dense, bilateral, reliable anesthesia (including sacral anesthesia), using a very low dose of drug with minimal risk of maternal toxicity or fetal drug transfer (table 3).

Greater reliability of spinal and CSE techniques was demonstrated in a systematic review that identified 54 randomized trials (3500 patients) that compared spinal or CSE with epidural anesthesia for elective CD [25]. Spinal and CSE together were associated with a lower rate of inadequate anesthesia compared with epidural anesthesia (10.2 percent [95% CI 9.0–11.4 percent], versus 30.3 percent [95% CI 26.5–34.5 percent]). The need for general anesthesia was very rare (0.1 percent), however, the incidence of breakthrough pain during surgery was high with all types of neuraxial anesthesia (14.7 percent). Management of inadequate neuraxial anesthesia during surgery is discussed below. (See 'Failed or inadequate neuraxial block' below.)

We use catheter neuraxial anesthesia techniques (ie, CSE, epidural, dural puncture epidural) rather than single-shot spinal when there is time to perform the procedure and achieve surgical anesthesia for the following situations (algorithm 1):

●Expected prolonged surgery, due to complicated repeat CD, extensive previous abdominal or uterine surgery, multiple procedures planned, high risk of hemorrhage, severe patient obesity, or cesarean hysterectomy.

●Patients with increased risks of adverse reactions or complications with general anesthesia (eg, predicted difficult airway, history of malignant hyperthermia, prior complications from general anesthesia).

●One contributor uses catheter neuraxial techniques for patients who have not fasted according to standard preoperative fasting guidelines or who have conditions other than pregnancy that increase the risk of aspiration. Others use a single shot spinal in this category of patients.

Some clinicians use dural puncture epidural (DPE) for anesthesia for CD, particularly in patients with obesity, to help verify the location of the epidural needle tip in the epidural space. For patients with obesity, the epidural needle can also be used as a longer introducer needle than the short introducer needle typically used for spinal anesthesia. (See "Neuraxial analgesia for labor and delivery (including instrumental delivery)", section on 'Dural puncture epidural analgesia'.)

If used for elective cesarean delivery, a DPE should be dosed the same way as a conventional epidural catheter. (See 'Epidural anesthesia' below.)

One study suggested that DPE may speed onset and improve block quality if the epidural is extended for surgical anesthesia, compared with conventional epidural. In this single institution trial, 136 patients scheduled for elective cesarean delivery were randomly assigned to receive DPE (using a 25 gauge spinal needle) versus conventional epidural placed approximately one hour prior to surgery, with epidural analgesia established and maintained with 0.0625% bupivacaine with fentanyl [26]. At the time of surgery, the epidural was extended with 3% chloroprocaine. The time to achieve a T6 sensory level was approximately three minutes faster and a composite outcome of block quality was better in patients who had DPE.

Spinal anesthesia — The techniques for single-shot and continuous spinal anesthesia (CSA) are described separately. (See "Spinal anesthesia: Technique".)

Spinal drugs for CD — A combination of local anesthetic (LA) and opioid is usually administered for spinal anesthesia for CD (table 4).

●Local anesthetics – Bupivacaine is the most commonly used LA for spinal anesthesia for CD in the United States because of its duration of action is appropriate for most CDs, a low incidence of transient radicular irritation, low cost, and wide availability. We prefer hyperbaric bupivacaine rather than isobaric bupivacaine because of its rapid onset and the option to modify the spinal level by changing the position of the operating table.

Other options include ropivacaine, levobupivacaine (outside the United States), and chloroprocaine. (See "Spinal anesthesia: Technique", section on 'Local anesthetics'.)

●Intrathecal opioids – Intrathecal opioids may be added to the LA solution to improve intraoperative analgesia and for postoperative analgesia.

•Lipophilic opioids (fentanyl and sufentanil) – Lipid-soluble opioids such as fentanyl and sufentanil are commonly added to intrathecal LA to improve intraoperative analgesia. They provide minimal postoperative analgesia because of their short duration of action. We routinely add fentanyl 10 to 15 mcg to bupivacaine for spinal anesthesia for CD. Sufentanil is less commonly used in the United States than in Europe because the concentrated solution (50 mcg/mL) makes accurate measurement of low doses difficult. Pruritus is a common dose-related side effect of intrathecal fentanyl and sufentanil [27].

A 2020 meta-analysis of 14 randomized trials (700 patients) of the effects of adding fentanyl to bupivacaine for spinal anesthesia for CD found that intrathecal fentanyl (10 to 15 mcg) reduced the need for supplemental intraoperative analgesia (4.5 percent versus 29.6 percent, risk ratio [RR] 0.18, 95% CI 0.11-0.27) and reduced the incidence of intraoperative nausea and vomiting (10 percent versus 25.8 percent, RR 0.41, 95% CI 0.24-0.74) [28]. The incidence of pruritus was increased sixfold with the addition of fentanyl. The results were similar when fentanyl was added to a combination of bupivacaine and morphine, based on two small trials. The optimal dose of intrathecal fentanyl could not be determined, though doses between 10 and 12.5 mcg reduced the need for intraoperative analgesia and resulted in low rates of intraoperative nausea and vomiting, and pruritus. The overall quality of data were judged to be moderate, based on a high risk of bias in some studies.

In another meta-analysis of 11 randomized trials that compared the use of fentanyl ≤12.5 mcg versus >12.5 mcg for spinal anesthesia for CD, the higher dose slightly reduced the need for intraoperative analgesia and prolonged the time for the first request for analgesia, but increased the incidence of pruritus [29]. There was significant statistical heterogeneity, and the quality of evidence was low to moderate.

•Hydrophilic opioids (morphine and hydromorphone) – Hydrophilic opioids (eg, preservative-free morphine or hydromorphone) are commonly added to LA for spinal anesthesia for CD, primarily for postoperative analgesia. Use of neuraxial opioids for post CD analgesia is discussed in detail separately. (See "Post-cesarean delivery analgesia", section on 'Neuraxial opioids'.)

●Adjuvants – Additional drugs that may be added to the spinal drug solution include the following:

•Epinephrine – Epinephrine (100 to 200 mcg intrathecal) may be added to intrathecal drugs to prolong the duration of action. The authors use 100 mcg of intrathecal epinephrine, which extends the surgical block time by 30 to 50 minutes [30].

•Clonidine – Intrathecal clonidine may be of benefit for spinal anesthesia in obstetric patients, though it is not approved by the US Food and Drug Administration (FDA) for spinal anesthesia. The addition of 50 to 75 mcg of intrathecal clonidine to hyperbaric bupivacaine improves early postoperative analgesia [31-33], while larger doses (150 mcg) may decrease postoperative hyperalgesia [34]. In one study, 24 hour postoperative opioid consumption after CD was similar in patients who received intrathecal clonidine and those who received intrathecal fentanyl [35].

One of the authors uses a low dose of intrathecal clonidine (15 mcg) in addition to 10 mcg of fentanyl, as this dose of clonidine has been associated with better hemodynamic and sedation profiles compared with higher doses (45 mcg) [36], and in our experience, decreases intraoperative visceral pain and shivering.

Clonidine is used by some clinicians in addition to or in place of intrathecal opioids in patients with a history of severe pruritus with opioids, substance use disorders, or chronic pain. (See "Obstetric anesthesia for patients with opioid use disorder or opioid tolerance", section on 'Patients who have neuraxial anesthesia'.)

•Neostigmine – The use of neostigmine in spinal anesthesia has been associated with significant increases in maternal nausea and vomiting and is not advised [37].

•Dexmedetomidine – Intrathecal dexmedetomidine has been used off-label as an adjunct to local anesthetics for neuraxial anesthesia for CD, primarily to prevent shivering [38-40], although neither author has used it. Its mechanism of action and effects are similar to those of epinephrine and clonidine [41] . However, rigorous neurotoxicity studies are lacking.

In a meta-analysis of four randomized trials that evaluated the effect of adding intrathecal dexmedetomidine (5 mcg) to bupivacaine for spinal anesthesia for CD, dexmedetomidine reduced intraoperative shivering (14 versus 34 percent, risk ratio 0.4, 95% CI 0.25-0.65) [40]. The incidences of intraoperative nausea and vomiting, bradycardia, and hypotension were similar. Intrathecal dexmedetomidine may also improve analgesia [38,39,42].

Drugs for continuous spinal anesthesia — The drugs used for CSA for CD are the same ones used for single-shot spinal. However, dosing must account for the dead space of the catheter (approximately 0.3 mL, but may vary) and may be reduced if the spinal catheter has been used for labor analgesia.

Importantly, inadvertent administration of epidural doses of medications through a spinal catheter can result in high or total spinal anesthesia, severe hypotension, and cardiac arrest. The doses of drugs used for CSA are approximately 10 percent of those that are used for epidural anesthesia. All members of the nursing, obstetric, and anesthesiology teams must be notified that the catheter is in the intrathecal space, and the catheter should be labeled appropriately. We dose spinal catheters for CD as follows:

●Spinal catheter placed de novo for CD – For a de novo spinal catheter, we incrementally administer the spinal bupivacaine, titrating to effect (anticipating the same total dose as for a single-shot spinal). A reasonable approach is to inject an initial dose of 5 to 7.5 mg bupivacaine (0.5% or 0.75%), titrated to desired level with 2.5 mg incremental boluses every three to five minutes [43].

We administer the intrathecal opioids and adjuvants as boluses to ensure adequate dosing.

●Spinal catheter initially placed for labor analgesia – When surgical anesthesia is necessary for patients with an in situ spinal catheter used for labor analgesia, the initial doses of LA for surgery must be reduced, depending on the labor analgesic infusion, and the patient’s sensory level. Spinal drugs should be administered slowly and titrated to effect in the operating room. Each labor floor should have a protocol for dosing spinal catheters for emergency CD.

For emergency CD in patients who are receiving continuous spinal labor analgesia, we administer 7.5 mg (1 mL) of 0.75% hyperbaric bupivacaine through the spinal catheter. This bolus typically results in adequate anesthesia for surgery; further doses of bupivacaine may be titrated in 1.5 mg (0.2 mL) increments, to effect. Since these patients have received lipophilic opioids (ie, fentanyl or sufentanil) as part of the infusion for labor analgesia, no additional lipophilic opioid is required for surgical anesthesia.

Intrathecal morphine or hydromorphone should be administered as a bolus for postoperative analgesia. (See "Post-cesarean delivery analgesia", section on 'Hydrophilic opioids (morphine and hydromorphone)'.)

When using an intrathecal catheter, we establish an operative level of anesthesia and then test sensory dermatomes roughly every hour to ensure adequate coverage. If the level is receding prior to the closure of the fascia, we administer incremental supplemental doses of spinal local anesthetic (eg, 0.5 mL of hyperbaric 0.75% bupivacaine [3.75 mg]) to maintain a T4-T6 level.

Epidural anesthesia — The technique for epidural catheter placement is discussed separately. (See "Epidural and combined spinal-epidural anesthesia: Techniques", section on 'Epidural anesthesia technique'.)

Initiation of epidural anesthesia — When dosing an epidural catheter for CD, local anesthetic should be titrated to effect, unless the CD is an emergency.

●Existing labor epidural – The total volume of LA required for conversion of epidural labor analgesia to surgical anesthesia varies with the concentration and dose of local anesthetic used for labor. Higher concentration labor analgesia solutions administered at a higher hourly rate may require less overall local anesthetic to convert to surgical anesthesia. As an example, with a low concentration background infusion (eg, bupivacaine ≤ 0.1% and no “top up” of more concentrated local anesthetic within the hour), conversion will likely require 15 to 20 mL of 2% lidocaine with epinephrine or 3% 2-chloroprocaine. In contrast, with a high concentration background infusion (eg, bupivacaine ≥0.125%) and/or a recent “top up” of more concentrated local anesthetic, conversion may require only 10 to 15 mL of local anesthetic.

For emergency CD, after aspiration for blood or cerebrospinal fluid (CSF), we inject up to 20 mL of LA in 5 mL increments in rapid succession, pausing between injections only to aspirate back on the catheter to ensure that there is no clear fluid or blood return.

●De novo epidural – After placing a de novo epidural catheter, achieving a surgical level of anesthesia may require up to 25 mL of local anesthetic.

Epidural drugs for CD

●Local anesthetics – The epidural LA used for CD include lidocaine, chloroprocaine, and less commonly, bupivacaine and ropivacaine (table 5).

•Choice of LA – The choice of epidural LA is usually based on the speed of onset required for the clinical situation and the possibility of systemic toxicity. Lidocaine 2% with epinephrine (1:200,000 dilution) is the most commonly used LA for CD. Bupivacaine and ropivacaine are rarely used for epidural anesthesia because of the much slower onset and higher risk of cardiotoxicity, compared with lidocaine and chloroprocaine.

Our practice is as follows:

-Scheduled – We typically use 2% lidocaine with epinephrine with or without bicarbonate.

-Urgent or emergency CD – We use bicarbonated 2% lidocaine with epinephrine or bicarbonated chloroprocaine 3%. The addition of sodium bicarbonate speeds the onset of lidocaine and chloroprocaine. In one study, alkalinization of epidural lidocaine with epinephrine for patients with in situ labor epidural catheters decreased the time from epidural bolus to surgical anesthesia for CD from 9.7 to 5.2 minutes [44].

The addition of sodium bicarbonate to 3% 2-chloroprocaine can decrease the onset of a rapid epidural bolus through an existing labor epidural catheter from approximately eight minutes [45] to approximately three minutes [46].

Prior to any rapid epidural injection, clinicians should aspirate the catheter for blood or CSF to assess for intrathecal or intravascular migration.

The optimal local anesthetic solution for rapid extension of epidural block is unclear. In a randomized trial including 67 patients with labor epidural analgesia in place, the time to onset of surgical epidural anesthesia was determined after a bolus of 2% lidocaine with epinephrine, bicarbonate, and fentanyl versus a bolus of 3% chloroprocaine [47]. The time from start of the epidural extension bolus to achieving a T7 sensory level was faster in patients who received the lidocaine solution, but was rapid with both chloroprocaine and lidocaine solutions (655 ± 258 and 558 ± 269 seconds, respectively) [47]. Use of chloroprocaine with fentanyl with or without bicarbonate was not evaluated.

The duration of action of 3% 2-chloroprocaine is approximately 40 to 60 minutes. Therefore, to maintain surgical anesthesia we administer 2% lidocaine with epinephrine 15 to 20 minutes (roughly one half-life) after the initial bolus of 2-chloroprocaine and an additional 5 to 10 mL of lidocaine as required at 35 to 50 minute intervals. In the rare case in which we need to continue using chloroprocaine (eg, in a patient with an amide local anesthetic allergy), we re-dose with 5 mL every 20 minutes throughout the duration of surgery.

Chloroprocaine may reduce the efficacy of analgesia from subsequently administered epidural morphine [48,49]. However, we still administer the epidural morphine towards the end of the CD and find it to be clinically effective.

●Epidural opioids – Opioids may be administered epidurally to speed the onset of anesthesia, improve the quality of the anesthetic, and/or provide postoperative analgesia [50-52]. Lipophilic opioids (fentanyl or sufentanil) are administered primarily for intraoperative analgesia and are administered during initiation of epidural anesthesia for CD. Hydrophilic opioids (ie, morphine and hydromorphone) are used for postoperative analgesia and are administered after the fetus is delivered. Doses for the commonly used opioids are shown in a table (table 5). (See "Post-cesarean delivery analgesia", section on 'Hydrophilic opioids (morphine and hydromorphone)'.)

●Other adjuvants – Additional drugs that may be added to the epidural solution include the following:

•Epinephrine – Epinephrine (1:400,000 to 1:200,000) is often added to the epidural local anesthetic solution to increase the density of block, reduce systemic absorption of LA, and prolong the duration of anesthesia [53]. (See "Epidural and combined spinal-epidural anesthesia: Techniques", section on 'Alpha-adrenergic agonists'.)

•Neostigmine – Epidural administration of neostigmine may reduce the required dose of LA for CD, and may modestly enhance postoperative analgesia, without significant maternal or fetal side effects [37]. Unlike intrathecal neostigmine, epidural administration does not seem to cause nausea and vomiting.

Neostigmine is not widely used as an adjunct for epidural anesthesia in the United States and is not used by the authors. Further study is required before recommending the routine use of neostigmine in this setting.

•Clonidine – Epidural clonidine is not widely used for CD, and there is very little literature on its use. One author of this topic typically administers 75 mcg of epidural clonidine in addition to epidural fentanyl as an adjunct to local anesthetic when converting a labor epidural to surgical anesthesia. (See "Epidural and combined spinal-epidural anesthesia: Techniques", section on 'Alpha-adrenergic agonists'.)

•Sodium bicarbonate – Sodium bicarbonate 8.4% may be added to the epidural local anesthetic solution (lidocaine or chloroprocaine) to speed the onset and improve the quality of the block (usually 2 mL/20 mL of local anesthetic, as described above). (See "Epidural and combined spinal-epidural anesthesia: Techniques", section on 'Sodium bicarbonate'.)

Combined spinal-epidural anesthesia — The technique for CSE is discussed separately. (See "Neuraxial analgesia for labor and delivery (including instrumental delivery)", section on 'CSE technique'.)

The drugs used for CSE mimic those used for spinal anesthesia and, if necessary, epidural anesthesia. Opioids are added to the spinal drugs (eg, fentanyl and preservative-free morphine). CSE can be initiated by injecting the same doses of spinal drugs as for single shot spinal, or by injecting a partial dose and then injecting the epidural catheter to achieve adequate anesthesia. The authors typically use a full dose spinal technique for CSE.

●Full spinal dose CSE – In most cases, the same initial doses and combinations of spinal drugs are administered as for single-shot spinal anesthesia. The first epidural dose is then administered when the sensory level recedes below T4. At that point, after negative aspiration of the catheter for blood and CSF, we administer a test dose of 3 mL of 1.5 or 2% lidocaine with epinephrine 1:200,000. In this setting, tachycardia with tinnitus or metallic taste may indicate an intravascular catheter. Intrathecal catheter placement may be more difficult to detect with the preexisting motor block. Unusually high levels (above C8) with the test dose should prompt consideration of intrathecal location of the catheter. (See "Neuraxial analgesia for labor and delivery (including instrumental delivery)", section on 'The epidural test dose in obstetrics'.)

If the initial spinal dose is adequate for surgery, epidural dosing or injection of even small volumes (eg, test dose) should not commence within the first 30 to 60 minutes after a full spinal dose was administered, as there is an increased risk of high spinal in this setting. If the initial spinal level is inadequate for surgery 15 minutes after the spinal injection, an epidural test dose may be administered sooner.

The test dose often reestablishes a level of surgical anesthesia above T4. We then test the sensory level every 20 minutes and administer aliquots of 3 to 5 mL of 2% lidocaine with epinephrine, based on the level of the block. (See "Epidural and combined spinal-epidural anesthesia: Techniques", section on 'CSE technique'.)

●Sequential CSE – For sequential CSE, a portion of the usual single-shot spinal dose is injected first, (eg, 5 to 6 mg bupivacaine), followed by epidural doses (eg, 3 to 5 mL of 2% lidocaine with epinephrine), titrated to achieve adequate surgical anesthesia.

Sequential techniques result in less hypotension [54-56] than single-shot or full-dose spinal CSE if the epidural medications are titrated slowly, and may indicate that the epidural catheter is well positioned at the start of the anesthetic.

Dural puncture epidural — Some clinicians use DPE for anesthesia for CD, particularly in patients with obesity, to help verify the location of the epidural needle tip in the epidural space. (See "Neuraxial analgesia for labor and delivery (including instrumental delivery)", section on 'Dural puncture epidural analgesia'.)

If used for elective cesarean delivery, a DPE should be dosed the same as a conventional epidural catheter. (See 'Epidural anesthesia' above.)

One study suggested that DPE may speed onset and improve block quality if the epidural is extended for surgical anesthesia, compared with conventional epidural. In this single institution trial, 136 patients scheduled for elective CD were randomly assigned to receive DPE versus conventional epidural, placed approximately one hour prior to surgery, with epidural analgesia established and maintained with 0.0625% bupivacaine with 2 mcg/mL fentanyl [26]. At the time of surgery, the epidural was extended with 3% chloroprocaine. The time to achieving a T6 sensory level was approximately three minutes faster and a composite outcome of block quality was better in patients who had DPE.

Intraoperative management

Intraoperative positioning — Patients can be positioned with left uterine displacement (LUD) for CD, to minimize the chance of aortocaval compression. (See "Cesarean birth: Preincision planning and patient preparation", section on 'Uterine displacement'.)

LUD has typically been recommended for any pregnant patient with a visibly gravid uterus or at greater than 20 weeks gestational age. The need for and efficacy of routine LUD have been questioned [57], and practice varies. One contributor to this topic routinely positions patients for CD with LUD whenever feasible, while another contributor uses LUD selectively, in patients with particular concerns for maintaining uteroplacental perfusion (eg, nonreassuring fetal heart rate trace, patients with preeclampsia, multiple gestation) or those who report feeling unwell lying supine. Tilting requires adequate support to keep the patient from sliding, and must maintain an adequate surgical field for the procedure.

Results of studies of the effect of LUD during spinal anesthesia on maternal hemodynamics are mixed, with some finding lower cardiac output and blood pressure with supine positioning, and others finding no difference [58-60].

Two randomized trials failed to demonstrate a difference in neonatal acidosis [58] or Apgar scores [59] with the use of LUD versus the supine position during spinal anesthesia for CD. Both studies used aggressive fluid administration and vasopressors infusions to maintain hemodynamic stability, and included only healthy parturients without signs of fetal compromise. Little is known about the effects of 15 degree tilt in parturients who may have uteroplacental insufficiency.

Magnetic resonance imaging (MRI) studies have found that an extreme degree of uterine displacement (ie, 30 degree tilt) may be required to reduce aortocaval compression [61-63]. This degree of tilt may be impractical during CD.

Supplemental oxygen — Most healthy parturients do not require supplemental oxygen during neuraxial anesthesia for uncomplicated CD. We administer supplemental oxygen via the anesthesia machine breathing circuit to parturients with worrisome preoperative fetal heart rate tracings or other evidence of fetal compromise, and to parturients with decreased oxygen saturation.

Preventing nausea and vomiting — Nausea with or without vomiting occurs commonly during CD with neuraxial anesthesia and may be due to hypotension, visceral manipulation (eg, uterine exteriorization or reinsertion), or administration of uterotonic medication. We routinely administer prophylaxis for intraoperative and postoperative nausea and vomiting (PONV). We administer ondansetron 4 mg intravenous (IV) for prophylaxis and for the possible hemodynamic benefits during spinal anesthesia [64,65] (see 'Ondansetron' below). Other options for PONV prophylaxis are discussed separately. (See "Postoperative nausea and vomiting", section on 'Antiemetics'.)

Hemodynamic management — Dense neuraxial blockade at the sensory level required for CD (ie, T4) causes sympathetic block and vasodilation, which can result in hypotension if not appropriately treated with a vasopressor (figure 1) [66].

Goal blood pressure — We aim to maintain blood pressure within 10 to 20 percent of baseline blood pressure unless the patient has severe hypertension. (See "Anesthesia for the patient with preeclampsia", section on 'Goal blood pressure'.)

We use a titrated prophylactic phenylephrine infusion as described below, with additional boluses of vasopressors to patients who drop below 20 percent of their baseline blood pressure, become symptomatic with nausea, vomiting, or lightheadedness, or if the fetus shows signs of distress following placement of spinal anesthesia. (See 'Choice and use of vasopressors' below.)

While recommendations are usually to maintain systolic blood pressure at ≥100 mmHg or ≥80 percent of baseline for healthy parturients, the optimal goal for blood pressure management has not been defined [67]. In one trial, approximately 75 parturients who had spinal anesthesia for CD were randomly assigned to have systolic blood pressure maintained at 80 percent, 90 percent, or 100 percent of baseline values, with titrated prophylactic phenylephrine infusion without fluid co-load [68]. Patients whose systolic blood pressure was maintained at 100 percent had significantly less nausea or vomiting and umbilical artery pH was higher compared with the other two groups, however none of the umbilical artery blood gases were acidotic.

Prevention and management of neuraxial anesthesia induced hypotension — Prophylaxis and/or timely treatment of hypotension may prevent maternal nausea and vomiting, fetal distress, and acidosis [69]. Here we discuss hemodynamic management of patients without preeclampsia. Use of vasopressors and hemodynamic management in patients with preeclampsia are discussed separately. (See "Anesthesia for the patient with preeclampsia".)

●Spinal induced hypotension – We use prophylactic infusion of phenylephrine, started at the time of spinal injection, supplemented as necessary with rescue boluses aiming for maternal blood pressure close to baseline. (See 'Goal blood pressure' above.)

This strategy is largely consistent with the 2017 International Consensus Statement on the Management of Hypotension with Vasopressors During Cesarean Section Under Spinal Anesthesia [70]. A network meta-analysis of 109 randomized trials involving various methods for the prevention of spinal hypotension at CD found that vasopressors were more effective than fluid loading, leg compression, or other types of medications [71].

The need for escalating doses of vasopressor during surgery or the inability to wean from a prophylactic infusion should prompt a multidisciplinary discussion and assessment of the degree of hemorrhage.

●Epidural induced hypotension – Prophylactic vasopressors are not typically needed during initiation of epidural anesthesia for non-urgent CD, as the onset of hypotension is delayed compared with spinal anesthesia. Therapeutic use of vasopressors during epidural anesthesia is similar to use of vasopressors during spinal anesthesia.

Choice and use of vasopressors — Phenylephrine, ephedrine, and norepinephrine have all been used to prevent and treat spinal hypotension in obstetrics. We suggest administration of phenylephrine rather than ephedrine, to prevent and treat neuraxial block-induced hypotension in the absence of maternal bradycardia. Ephedrine has been associated with a slight increase in fetal acidosis, though of questionable clinical significance. Norepinephrine is a reasonable alternative vasopressor for prophylaxis and treatment of spinal hypotension and is increasingly appearing in the literature. Characteristics and doses of these vasopressors appear in a table (table 6).  

●Phenylephrine – Phenylephrine is a direct alpha-1 adrenergic agonist, and as such can cause reflex bradycardia.

Prophylactic infusion of phenylephrine to prevent spinal hypotension is associated with less overall maternal nausea and vomiting compared with bolus rescue administrations [72-74]. However, there is no known difference in fetal acidemia between the use of prophylactic versus rescue dosing of phenylephrine [75].

We start a low dose prophylactic infusion at 25 to 75 mcg/minute IV at the time of injection of the spinal drug. As the spinal takes effect, we titrate the infusion upwards as high as 100 mcg/minute IV, adding supplemental boluses of 50 to 100 mcg IV if necessary to maintain blood pressure at or near baseline. (See 'Goal blood pressure' above.)

Prophylactic phenylephrine infusions at high doses (75 to 100 mcg/minute) are associated with a higher incidence of bradycardia and hypertension [76-80]. Profound bradycardia (40 to 50 beats per minute) without hypotension should be treated by reducing or stopping the infusion. Bradycardia with associated hypotension can be treated with vasopressors that have beta-1 activity including ephedrine, norepinephrine or epinephrine, and may require the use of an anticholinergic medication (glycopyrrolate, 0.2 to 0.4 mg IV).

●Ephedrine – Ephedrine is an indirect sympathomimetic, acting at alpha-1, beta-1 and beta-2 adrenergic receptors. As such, it produces increases in both blood pressure and heart rate.

When used for treating spinal hypotension during CD, compared with phenylephrine, ephedrine causes a slight increase in fetal acidosis, (though of questionable clinical significance), without a difference in Apgar scores [71]. A postulated mechanism for acidosis is increased fetal metabolic rate, as ephedrine crosses the placenta [81]. In addition, ephedrine has a delayed onset compared with phenylephrine, which may result in a longer period of hypotension.

Most of the studies comparing ephedrine and phenylephrine for spinal hypotension included only healthy parturients undergoing scheduled CD. Thus, it is unclear which vasopressor would be preferable in cases of non-reassuring fetal heart rate trace. In one retrospective study of approximately 380 high-risk parturients who underwent CD under spinal anesthesia, umbilical artery pH was similar in patients who received ephedrine, phenylephrine, or required no vasopressor during the anesthetic [82]. Conclusions are limited by the retrospective nature of the study and lack of information on other aspects of management.

●Norepinephrine – Norepinephrine is a direct acting sympathomimetic with effects at alpha-1 and -2 and beta-1 and -2 adrenergic receptors. Thus, it increases blood pressure and may maintain heart rate. It may be preferred for prophylaxis or treatment of patients with bradycardia, as an alternative to ephedrine [76,83-85].

We do not routinely use norepinephrine to prevent or treat spinal anesthesia induced hypotension. Existing data are limited, results are mixed, and there are insufficient evidence to recommend using norepinephrine rather than phenylephrine.

If norepinephrine is used, a reasonable starting dose would be 2.5 to 4 mcg/minute, titrated to effect [86]. Importantly, before administration of norepinephrine, patency of the intravenous catheter and appropriate pump settings should be verified, as extravasation can result in significant tissue damage and overdose can have devastating effects.

Most (not all) studies that have compared norepinephrine with phenylephrine have found less bradycardia in patients who receive norepinephrine, with similar fetal acidosis and Apgar scores [83,85,87-91].

IV fluid bolus — For most patients, we suggest administering a rapid bolus of IV crystalloid at the time of induction of spinal anesthesia (co-loading), in conjunction with prophylactic and as needed vasopressors. We typically administer a fluid bolus of 500 to 1000 mL IV during the neuraxial procedure, depending on the patient's comorbidities and volume status [66]. Further fluid administration should be based on an accurate estimate of blood loss. In patients with preeclampsia with severe features, a limited fluid bolus should be administered. (See "Anesthesia for the patient with preeclampsia", section on 'Intravenous fluid management'.)

Crystalloid solutions are used more commonly than colloid solutions for CD because they are less expensive and more readily available. Furthermore, the available data do not indicate an absolute benefit of colloids over crystalloids [92-96].

Glucose-free solutions should be used to prevent hypoglycemia in the fetus. Excessive placental glucose transfer can result in compensatory release of fetal insulin (fetal hyperinsulinemia), resulting in neonatal hypoglycemia after delivery [97,98].

Ondansetron — Ondansetron administered at the time of spinal injection may reduce the incidence of spinal hypotension and related symptoms. Two meta-analyses of randomized trials reported that ondansetron administered prophylactically during spinal anesthesia for CD reduced the incidence of hypotension, the dose of vasopressors administered, and the incidence of bradycardia, although the included studies were small with high heterogeneity [64,65]. A randomized trial including 60 patients who underwent elective CD with combined spinal epidural anesthesia found that a single dose of ondansetron 4 mg IV reduced the ED₅₀ (ie, effective dose in 50 percent of patients) of phenylephrine infusion by approximately 26 percent [99]. In this study, phenylephrine was administered to maintain blood pressure within 20 percent of baseline.

Maintaining normothermia — Similar to other patients having surgical procedures, patients who undergo CD are at risk for hypothermia due to prolonged skin exposure and fluid shifts.

We routinely attempt to maintain normothermia in both the mother and the neonate by keeping the room temperature at 22 to 23°C and using warmed IV fluids if available. We cover the mother with warm blankets for routine cesarean deliveries with neuraxial anesthesia of less than an hour duration and use an active warming device (eg, forced-air warming) for longer procedures and for patients who have general anesthesia.

We routinely monitor temperature for patients who have general anesthesia or blood transfusion, and for patients who have neuraxial anesthesia for longer procedures. In non-obstetric patients, intraoperative hypothermia is associated with coagulopathy, infection, prolonged drug effects, and myocardial ischemia, though these outcomes have not been studied in obstetric patients. (See "Perioperative temperature management", section on 'Consequences of hypothermia'.)

The optimal methods (eg, prewarming and active intraoperative forced air warming, warmed fluids, increased ambient temperature) for preventing hypothermia during CD are unclear; available literature involves varied interventions and duration of surgery [77,78].

The mechanisms for perioperative hypothermia, consequences, and preventive measures are discussed in detail separately. (See "Perioperative temperature management".)

Neonatal effects of hypothermia are also discussed separately. (See "Neonatal resuscitation in the delivery room", section on 'Temperature control'.)

Treating shivering — Shivering occurs in up to 50 percent of patients who receive spinal anesthesia for cesarean delivery [100]. For patients who require treatment, we administer dexmedetomidine 20 to 30 mcg IV.

A network meta-analysis of 20 randomized trials (1983 patients) that compared intravenous treatments for shivering during CD identified four effective treatments [101]. Dexmedetomidine was the most effective treatment, followed in decreasing order of efficacy by tramadol, nalbuphine, and meperidine. Dexmedetomidine was also the treatment least likely to cause maternal nausea.

Administration of uterotonics — Uterine contraction is the primary mechanism for reduction of uterine bleeding after delivery. The uterus is massaged, and oxytocin is administered as the first-line uterotonic medication.

We suggest administration of oxytocin by controlled infusion, with or without a 1-unit bolus for elective cesarean delivery, or a 3-unit diluted bolus (given over >30 seconds) for an intrapartum cesarean delivery. The lowest effective dose of oxytocin should be used, as the incidence of side effects is directly related to the rate of administration. Rapid or high dose administration of oxytocin can cause hypotension, nausea and vomiting, and chest pain with electrocardiogram changes [79], and can cause cardiovascular collapse [80,102]. Each labor unit should have a standardized protocol for oxytocin administration. Oxytocin has a narrow therapeutic window and is a high-alert medication according to the Institute of Safe Medication Practices [103]. Protocols vary by institution; an example of one such protocol is shown in an algorithm (algorithm 2). In general, oxytocin infusion is started at the time of cord clamping and titrated thereafter based on ongoing assessment of uterine tone.

We start oxytocin infusion at higher doses (eg, 7.5 to 15 units per hour [0.125 to 0.25 units per minute]) for patients who are at higher risk of uterine atony (eg, patients with prolonged labor, who received oxytocin during labor, with chorioamnionitis, multiple gestation).

●Patients who receive oxytocin during labor – Parturients who receive oxytocin during labor may become desensitized and require higher doses of oxytocin for a uterotonic effect. As an example, in a study of 30 patients who underwent CD after a mean of 10 hours of oxytocin during labor, the minimum effective bolus dose of oxytocin required to produce adequate uterine tone in 90 percent of patients (ED90) was 2.99 units (95% CI 2.32-3.67 units) [104]. In contrast, in a separate study, the ED90 of oxytocin for patients who underwent elective CD was approximately 9 fold less, 0.35 units (95% CI 0.18-0.52 units) [105]. Similarly, in other studies the ED90 for oxytocin infusion was approximately 15 units/hour for elective cesarean deliveries but was 44 units/hour in previously laboring patients with oxytocin supplementation during labor [106,107].

●Patients with increased uterine distention – Patients with greater than average uterine distention (eg, multiple gestation, polyhydramnios) are at increased risk of uterine atony, and may require higher doses of oxytocin and additional uterotonic drugs. As an example, in a study of 30 patients with twin pregnancies who underwent CD, the ED90 for the bolus dose of oxytocin was 4.38 units IV, and seven patients required rescue uterotonic drugs [108].

Patients with obesity may also require higher doses of oxytocin to achieve satisfactory uterine tone during CD. Whether the initial doses should be modified for patients with various degrees of obesity is unclear. In a dose finding study of the required oxytocin infusion rate at the time of CD in patients with and without obesity, the ED90 for prophylactic oxytocin infusion was greater in patients with BMI ≥30 kg/m² compared with patients with BMI <30 kg/m² (25.7 versus 16.6 units per hour, [relative ratio 1.55, 95% CI 1.1–2.0]) [109]. The need for additional uterotonic medications and estimated blood loss were similar in the two groups. Average weights in the two groups were 34.4 and 24.2 kg/m², respectively.

Other uterotonic medications and medications that are used to manage postpartum hemorrhage are discussed separately. (See "Postpartum hemorrhage: Medical and minimally invasive management", section on 'Administer additional uterotonic medications'.)

Failed or inadequate neuraxial block

Inadequate initial block — For elective CD, the level of sensory block should be assessed before skin preparation and draping to allow for supplementation or repeat neuraxial block if necessary. Management of inadequate block (no, low, or patchy block) depends on the degree of urgency and patient factors. Surgery should never be allowed to proceed with an inadequate block; IV analgesic supplementation alone is not appropriate [110].

●Inadequate or partial spinal block – If time permits, adequate time should be allowed for full onset of the spinal block before further intervention. If after 10 minutes there is a low or patchy spinal block, we usually place an epidural catheter and titrate the epidural medication to effect, similar to a sequential combined spinal epidural. (See 'Combined spinal-epidural anesthesia' above.)

Repeat single-shot spinal should be avoided, as the correct dose of the drug in this scenario cannot be determined, and either high spinal or inadequate spinal is possible. Some providers will perform a low dose CSE in this setting. However, there is a risk of high spinal if this is done. We prefer to place an epidural catheter which can be dosed precisely to minimize the risk of high spinal. For urgent CD, without time for epidural placement and dosing, general anesthesia should be considered.

●No spinal block – If after 20 minutes there is no evidence of spinal anesthesia, it is likely safe to repeat full dose spinal anesthesia.

●Inadequate epidural anesthesia – If epidural anesthesia is inadequate for CD after administration of appropriate doses of epidural medications, we replace the epidural catheter and titrate epidural medication to effect with the new catheter. Single-shot spinal should not be performed following failed epidural anesthesia if there is evidence of motor or sensory block. High block associated with single-shot spinal after failed epidural is one of the most common serious complications of obstetric anesthesia and can result in respiratory arrest [23] (see 'High neuraxial block' below). Some clinicians use a CSE in this setting with a low dose of spinal medication, though there are no data to guide the optimal dose of spinal medication [25,111].

Pain during surgery — Pain during neuraxial anesthesia for CD is common. In a 2022 systematic review of the literature on inadequate neuraxial anesthesia during elective CD, 14.6 percent of patients required conversion to general anesthesia or supplemental analgesia [25].

If the patient has moderate to severe intraoperative pain, if possible the procedure should be paused to assess the block and manage the pain. The use of neuraxial and intravenous adjuncts should be initiated based on the clinical situation [112]. Further management should be discussed with the obstetrician and with the patient. All intraoperative pain should be treated, and if necessary, general anesthesia should be induced.

The sensory level should be checked. Further management should be as follows:

●Receding or inadequate sensory level – If an epidural catheter is in place, re-dose the catheter with local anesthetic and epidural adjuncts (eg, fentanyl, clonidine). If there is no catheter in place, options depend on the time left in the procedure and the degree of pain and include IV opioid, IV dexmedetomidine, local anesthetic infiltration, intraperitoneal local anesthetic instillation, or induction of general anesthesia.

●Visceral discomfort with an adequate sensory level – If the uterus has been exteriorized, the possibility of internalizing it should be discussed with the obstetrician. Other options include IV opioids, intraperitoneal installation of local anesthetic, and administration of epidural adjuncts (eg, fentanyl, clonidine) if a catheter is in place.

For intraperitoneal local anesthetic instillation, chloroprocaine is often used to minimize the risk of local anesthetic systemic toxicity (LAST). At one contributor’s institution, the surgeons apply a sterile solution of 20 mL of 3% 2-chloroprocaine diluted with 20 mL of saline directly into the peritoneal cavity or onto the fascia. The optimal volume and concentration of local anesthetic solution for this technique have not been determined, and the incidence of LAST is unknown. In one case series of 32 patients who had pain during CD, 20 to 60 mL of 1% chloroprocaine (mean dose 11.8 mg/kg) was poured into the peritoneal cavity after delivery of the fetus [113]. One to five minutes later, the excess LA solution was suctioned away. Approximately one-half of the patients had improved pain scores after intraperitoneal LA, though all patients also received intravenous opioids. None of the patients exhibited signs of LAST.

High neuraxial block — High thoracic levels of spinal or epidural blockade often result in dyspnea. For most patients, reassurance that this is a normal sensation is sufficient. Assessment of grip strength is a quick way to determine whether the block involves the low cervical & high thoracic nerve roots and may be reaching a level that requires ventilatory support. Patients with weak grip strength should be assessed continuously for block progression. In the case of ascending spinal anesthesia with hyperbaric solution, the head of the bed should be elevated to prevent further cephalad spread. If the patient is unable to phonate or to maintain adequate oxygenation, general anesthesia should be induced.

High neuraxial block can rarely result in respiratory compromise, hypoxemia, hypotension, and cardiac arrest. The reported incidence of high neuraxial block in contemporary obstetric anesthesia ranges from approximately 1 in 4300 to one in 11,000 neuraxial anesthetics, based on case series and registry studies [114]. As an example, the Serious Complications Repository Project from the Society for Obstetric Anesthesia and Perinatology collected data on over 250,000 obstetric anesthetics [23]. High neuraxial block requiring intubation was the most frequently reported anesthesia-related complication and occurred in 1/4336 anesthetics. In this and in other studies, high blocks were associated with both unrecognized intrathecal catheters and with de novo spinal placement following failed epidural anesthesia [115]. Clinicians should always aspirate epidural catheters prior to injection and should inject medication slowly and incrementally if the clinical situation permits. In addition, if an epidural catheter fails to adequately provide anesthesia after dosing for CD, spinal anesthesia should be avoided as described above. (See 'Failed or inadequate neuraxial block' above.)

Management of high neuraxial block is discussed separately. (See "Overview of neuraxial anesthesia", section on 'High or total spinal anesthesia'.)

GENERAL ANESTHESIA — 

For patients who have general anesthesia (GA), patient positioning, hemodynamic management, temperature control, and administration of uterotonic agents, are the same as they are during neuraxial anesthesia. (See 'Intraoperative management' above and 'Administration of uterotonics' above.)

Patients who have general anesthesia for CD are at increased risk of difficulty with airway management, and of awareness with recall after anesthesia, compared with non-pregnant patients who undergo other types of surgery.

●Airway concerns – The incidence of airway difficulty and management of the airway for these patients are discussed separately. (See "Airway management for the pregnant patient".)

●Awareness during general anesthesia – Awareness with recall is relatively common during GA for CD compared to other surgeries, with rates as high as 1:212 cases, compared with 1:19,600 for all types of surgery [116,117]. The 5^th National Audit Project from the United Kingdom reported that many of these cases of awareness were preventable [116].

General anesthesia during CD is often administered differently than for other surgeries due to fetal concerns and maternal physiology; some of the changes in anesthetic technique may contribute to the high incidence of awareness in obstetric patients. Suggested mechanisms for awareness in these patients include:

•Underdosing or omission of sedative premedication, induction agents, and opioids to avoid placental transfer to the fetus

•Underdosing of inhalation anesthetics to avoid uterine relaxation and bleeding

•Changes in drug distribution and alveolar concentrations of anesthetic gases related to physiologic changes of pregnancy (eg, increase in cardiac output)

•Increase in difficulty with airway management, which is associated with awareness in all patients

•High percentage of emergencies among patients who have general anesthesia for CD (see "Accidental awareness during general anesthesia")

To minimize the possibility of awareness, we administer a high concentration of volatile anesthetic and/or additional propofol after induction of anesthesia leading up to delivery, and routinely administer midazolam and fentanyl at cord clamping and delivery. (See 'Maintenance of general anesthesia' below.)

Preparation for general anesthesia — Preoperative assessment and preparation for anesthesia are similar whether neuraxial anesthesia or general anesthesia is used. (See 'Preoperative assessment' above and 'Preparation for anesthesia' above.)

Prior to general anesthesia, after the patient is positioned, the abdomen is typically prepared and draped prior to induction of anesthesia, to minimize fetal exposure to anesthetics during the time between induction and delivery. The surgeon makes an incision only after the anesthesia clinician has the airway secured and/or indicated that it is appropriate to proceed.

Induction of anesthesia — Rapid sequence induction and intubation (RSII) has been the standard induction technique for CD. Some guidelines have suggested including low pressure mask ventilation to optimize oxygenation and confirm the feasibility of mask ventilation [118]. RSII and airway management for pregnant patients are discussed separately. (See "Airway management for the pregnant patient" and "Rapid sequence induction and intubation (RSII) for anesthesia".)

●Induction agents – Propofol (2 to 2.5 mg/kg intravenous [IV]) or etomidate (0.3 to 0.5 mg/kg IV) are the induction agents of choice, although ketamine (1 to 2mg/kg) may be used for hemodynamically unstable or hypovolemic patients. All intravenous induction medications are lipophilic and therefore transfer across the placenta to the fetus. The choice of induction agent should be determined by maternal factors. Since parturients typically do not receive premedication and have an increased volume of distribution compared with nonpregnant patients, full induction doses based on actual body weight should be administered for general anesthesia. (See "General anesthesia: Intravenous induction agents".)

●Adjunctive medications for induction – For most patients, we avoid administration of intravenous opioids until after the fetus is delivered, though practice varies. In a meta-analysis of 17 randomized placebo controlled trials of the use of alfentanil, remifentanil, or fentanyl during induction of anesthesia for CD, there were no significant differences in Apgar scores at one or five minutes after administration of alfentanil or remifentanil [119]. Based on four studies, fentanyl caused a likely clinically irrelevant reduction in five-minute Apgar scores (weighted mean difference [WMD] -0.2).

We administer adjunctive medications, including opioids and vasoactive medications, as indicated by the patient’s medical status. For patients with preeclampsia, opioids and other medications are administered during induction of anesthesia to prevent hypertension with laryngoscopy and intubation. (See "Anesthesia for labor and delivery in high-risk heart disease: General considerations" and "Anesthesia for the patient with preeclampsia".)

●Neuromuscular blocking agents (NMBAs) – NMBAs do not cross the placenta and are therefore not associated with weakness in the neonate. For routine CD, NMBAs are not required for surgery beyond endotracheal intubation. If further neuromuscular block is required to facilitate surgery, as for all surgical patients, dosing and reversal should be guided by neuromuscular monitoring, and every effort should be made to avoid residual neuromuscular block (see "Clinical use of neuromuscular blocking agents in anesthesia" and "Monitoring neuromuscular blockade"). Reversal of NMBA may safely be achieved with either neostigmine and glycopyrrolate or sugammadex as needed.

There are little data on the safety of sugammadex in patients who will breastfeed. Sugammadex is a large, polarized molecule and is therefore unlikely to transfer into breast milk in a significant quantity. A 2019 statement from the Society of Obstetric Anesthesia and Perinatology (SOAP) recommended caution when using sugammadex during CD [120], while the Lactmed database designates sugammadex acceptable to use during breastfeeding [121]. Consistent with 2023 guidelines from the American Society of Anesthesiologists, both authors would consider using sugammadex when necessary in patients undergoing CD (table 7). (See "Clinical use of neuromuscular blocking agents in anesthesia", section on 'Reversal of neuromuscular block'.)

Maintenance of general anesthesia — After induction and intubation, we either administer high-dose volatile anesthetic (eg, sevoflurane) with high flows of oxygen to quickly obtain an adequate end-tidal concentration of anesthetic (1 MAC) or additional propofol to minimize the chance of awareness under anesthesia.

Volatile inhalation anesthetics produce dose-dependent uterine relaxation [122], which can result in uterine atony and hemorrhage. After delivery, once fetal transfer of medication is no longer a concern, we administer midazolam (2 mg), a short-acting opioid (eg, fentanyl 100 mcg IV), and either N₂O 50 to 70 percent or a propofol infusion, to allow a reduced dose of volatile anesthetic to 0.5 MAC.

Ventilation should be controlled or supported to achieve an end-tidal carbon dioxide of approximately 30 mmHg, which should correlate with the normal partial pressure of carbon dioxide (PCO₂) of 30 to 32 mmHg at the end of pregnancy. (See "Anesthesia for nonobstetric surgery during pregnancy", section on 'Mechanical ventilation'.)

Emergence from anesthesia — Airway complications in obstetric patients are more common during emergence from anesthesia and the immediate postoperative period than during induction and initial airway management [123]. Emergence and extubation in these patients are discussed separately. (See "Airway management for the pregnant patient", section on 'Extubation'.)

POST-CESAREAN DELIVERY ANALGESIA — 

Multimodal strategies for postoperative pain control after CD should be used to promote rapid recovery, allow the patient to care for their newborn, and minimize the need for postoperative opioids. For most patients, the strategy for post-CD pain control should consist of neuraxial opioids and scheduled nonopioid analgesics (acetaminophen and nonsteroidal anti-inflammatory drugs [NSAIDs]), with systemic opioids reserved for breakthrough pain. Alternative plans are required for patients who receive general anesthesia for CD and may be required for patients with risk factors for severe postoperative pain (eg, patients with chronic pain and/or who chronically use opioids). Post-CD analgesia is discussed in detail separately. (See "Post-cesarean delivery analgesia".)

ENHANCED RECOVERY AFTER CESAREAN DELIVERY — 

Enhanced recovery protocols have been developed for CD, similar to other types of surgery. Such protocols consist of multidisciplinary best practices and include components that are used throughout the perioperative period [124-126]. The optimal elements of enhanced recovery after surgery (ERAS) have not been determined, and protocols are institution-specific. Our ERAS protocols are similar to the enhanced recovery after cesarean (ERAC) protocol published by the Society for Obstetric Anesthesia and Perinatology [127]. Common anesthesia-related components of ERAC protocols are shown in a table (table 8).

SOCIETY GUIDELINE LINKS — 

Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Local anesthetic systemic toxicity" and "Society guideline links: Post dural puncture headache" and "Society guideline links: Obstetric anesthesia" and "Society guideline links: Enhanced recovery after surgery".)

INFORMATION FOR PATIENTS — 

UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Anesthesia for cesarean birth (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Preoperative assessment and preparation

•Parturients undergoing cesarean delivery (CD) should have a focused history and physical examination by an anesthesia clinician. Antenatal anesthesia consultation should be considered for patients at risk for complications due to preexisting conditions (table 1). (See 'Preoperative assessment' above.)

•For patients who undergo elective or emergency CD, pharmacologic aspiration prophylaxis should be considered. Options are shown in a table (table 2). (See 'Preparation for anesthesia' above.)

●Choice of anesthetic technique – The choice of anesthetic technique for CD should be based on maternal and fetal status, comorbidities, expected duration and difficulty of the procedure, and the presence (or absence) of an in situ epidural catheter. If a working epidural catheter is in place, the epidural catheter is used to achieve surgical anesthesia (algorithm 1).

For most patients, we strongly prefer neuraxial anesthesia rather than general anesthesia (GA) for CD. There are exceptions for which GA is more appropriate. While maternal and neonatal mortality is not clearly improved with neuraxial anesthesia, most patients prioritize being awake for the birth of the infant.

Neuraxial anesthesia is associated with reduced risks of venous thromboembolism, surgical site infection, and severe postpartum depression. (See 'General versus neuraxial anesthesia' above.)

●Neuraxial anesthesia

•For patients without a labor epidural catheter in place and for whom neuraxial anesthesia is appropriate, we suggest using a spinal or combined spinal epidural (CSE) rather than a standard epidural anesthetic (Grade 2C). Spinal anesthesia is usually performed more quickly than epidural anesthesia and provides rapid onset of dense, bilateral, reliable anesthesia, using a very low dose of drug with minimal risk of maternal toxicity or fetal drug transfer. (See 'Choice of neuraxial technique' above.)

•For spinal anesthesia, the spinal solution typically includes local anesthetic (LA; eg, bupivacaine), lipophilic opioid (eg, fentanyl) to improve intraoperative anesthesia and hydrophilic opioid (eg, preservative-free morphine) for postoperative analgesia (table 4). (See 'Spinal drugs for CD' above.)

•For epidural anesthesia, the epidural solution typically includes LA (eg, 2% lidocaine), lipophilic opioid (eg, fentanyl) to improve intraoperative anesthesia, and hydrophilic opioid (eg, preservative-free morphine) for postoperative analgesia (table 5). For emergency CD, bicarbonated 3% 2-chloroprocaine or bicarbonated 2% lidocaine provide similarly rapid onset of surgical anesthesia. (See 'Epidural drugs for CD' above.)

•Management of failed or inadequate neuraxial anesthesia depends on the urgency of the situation and patient factors. For emergency CD, general anesthesia should be induced. When time permits, neuraxial anesthesia can be performed, taking precautions to avoid high neuraxial block. (See 'Failed or inadequate neuraxial block' above and 'High neuraxial block' above.)

-Inadequate spinal anesthesia – If possible, at least 10 minutes should be allowed after spinal injection to assess full onset. For partial block, we typically place an epidural catheter and titrate to effect. If there is no evidence of block, it is likely safe to repeat full dose spinal.

-Inadequate epidural anesthesia – We replace the epidural catheter and titrate epidural medication to effect. Single-shot full dose spinal should not be performed following failed epidural anesthesia, to avoid high neuraxial block.

●General anesthesia – For general anesthesia, our usual strategy is as follows:

•We perform rapid sequence induction and intubation. (See 'Induction of anesthesia' above.)

•After induction, we administer high concentrations of volatile anesthetic (eg, sevoflurane,) to rapidly achieve 1 MAC end-tidal anesthetic and minimize the risk of awareness during anesthesia. (See 'Maintenance of general anesthesia' above.)

•After delivery, we administer opioid (eg, fentanyl 100 mcg intravenous [IV]), midazolam 2 mg IV, and nitrous oxide 50 to 70 percent, and reduce the concentration of volatile anesthetic to 0.5 MAC or substitute a propofol infusion to reduce the risk of uterine atony and hemorrhage.

●Patient positioning – Patients should be positioned with left uterine displacement if feasible, as some will experience supine hypotension. (See 'Intraoperative positioning' above.)

●Hemodynamic management – In the absence of maternal bradycardia, we suggest using phenylephrine, rather than ephedrine, to prevent and treat neuraxial block-induced hypotension (Grade 2C). Compared with phenylephrine, ephedrine is associated with a slight increase in fetal acidosis, although of questionable clinical significance. Low dose norepinephrine may be a reasonable alternative for prevention and treatment of spinal hypotension (table 6).

For healthy patients, we administer a prophylactic, low dose, titrated infusion of phenylephrine with phenylephrine rescue boluses, along with intravenous volume expansion (co-loading) with glucose free crystalloid solution, aiming for maternal blood pressure within 20 percent of baseline, and aiming for lack of nausea and vomiting. (See 'Hemodynamic management' above.)

●Administration of uterotonics – In addition to uterine massage, oxytocin is the first line uterotonic medication. We suggest the administration of oxytocin by controlled infusion, with or without a 1-unit bolus for elective cesarean delivery, or a 3-unit diluted bolus (given over >30 seconds) for an intrapartum cesarean delivery (Grade 2C) (algorithm 2).

The lowest effective dose of oxytocin should be used, as the incidence of side effects (eg, hypotension, nausea, vomiting, chest pain, cardiovascular collapse) is directly related to the rate of administration. (See 'Administration of uterotonics' above.)