Postcesarean pain has both somatic and visceral components. Somatic pain arises from nociceptors within the abdominal wound and has both cutaneous and deep components, which are transmitted within the anterior divisions of the spinal segmental nerves, usually T10–L1. These nerve fibers run laterally in the abdominal wall between the layers of the transversus abdominis and internal oblique muscles (12). Visceral uterine nociceptive stimuli return via afferent nerve fibers that ascend through the inferior hypogastric plexus and enter the spinal cord via the T10-L1 spinal nerves (13,14). An ideal postcesarean analgesic regimen would be one that is cost-effective, simple to implement, and has minimal impact on staff workload. It would provide consistent and high-quality pain relief while catering to wide interpatient variabilities, and would also have a low incidence of side effects and complications. This ideal regimen would not interfere with the maternal care of the newborn or with breastfeeding, and there would be minimal drug transfer into the breast milk and consequently minimal adverse effects on the newborn. Achieving these goals requires a multimodal approach (15,16).

The safety benefits of regional anesthesia over general anesthesia in the pregnant patient are well documented (21). Most cesarean deliveries are performed using spinal, epidural, or combined spinal–epidural anesthesia (CSE) techniques. These methods also provide a convenient and effective route for neuraxial opioid administration which augment intraoperative anesthesia and optimize postoperative analgesia. Potency, onset, duration of action, and side effects vary depending on the opioid used and the route of its administration. Pruritus, nausea, and vomiting are the most common side effects of these agents and cause a decrease in maternal satisfaction.

Neuraxial opioids have the advantage of producing analgesia without motor or sympathetic blockade. Onset of analgesia is more rapid with the highly lipid soluble opioids. Conversely, lipid insoluble opioids, such as morphine, are retained in the cerebrospinal fluid (CSF), providing a longer supply to the spinal cord and consequently a slower onset, but longer duration of analgesia after the administration of a single dose (22).

Lipophilicity, as assessed by octanol/buffer distribution coefficient, does correlate with the meningeal permeability coefficient but in a nonlinear fashion. The optimal octanol/buffer distribution coefficient that results in maximal meningeal permeability is between 129 (alfentanil) and 560 (bupivacaine) (22). This biphasic relationship between lipophilicity and a drug’s meningeal permeability coefficient, may be explained by the dual nature of the arachnoid membrane which is the main barrier. After a drug is deposited in the epidural space but before it reaches the spinal cord, it must first cross a hydrophilic zone (extracellular and intracellular fluids) and then a hydrophobic zone (cell membrane lipids) of the arachnoid membrane (22). Consequently before diffusion through these two areas occur, the drug must first dissolve in those environments. Lipophilic drugs (i.e., those with high octanol/buffer partition coefficients, e.g., fentanyl and sufentanil) readily dissolve in the lipophilic component of arachnoid mater and can cross the region easily. Conversely, they penetrate the hydrophilic zone with difficulty creating the rate-limiting factor in their diffusion through the arachnoid membrane. Drugs with intermediate lipophilicity move more readily between the lipid and the aqueous zones, and their meningeal permeability coefficients are correspondingly greater (e.g., alfentanil, hydromorphone, meperidine) (22). These physical and chemical properties of the opioids will also determine vascular permeability. Opioids with high octanol/buffer distribution coefficients, such as fentanyl and sufentanil move more easily to the intravascular compartment than to the subarachnoid compartment. In this way spinal cord concentrations of an opioid following epidural administration are the result of the net difference between the rate of uptake and distribution to the vascular and subarachnoid spaces. These differences explain why morphine, despite having a meningeal permeability coefficient similar to fentanyl and sufentanil, which are well below the optimal range of meningeal penetration, is a useful agent for epidural analgesia.

Bernards and Hill also showed with their in vitro model that the octanol/buffer distribution coefficient for sufentanil was beyond the range of optimal meningeal permeability.

Respiratory depression somnolence and pruritus appear to be associated with the degree of rostral migration of the opioid in the CSF (8,22,23). The timing for the appearance of these side effects varies between lipophilic and hydrophilic opioids after epidural administration. Morphine’s rostral migration is a phenomenon which is dose dependent and follows a predictable time course (23). In contrast, rostral migration as determined by the onset of upper body analgesia (24) and the incidence of respiratory depression after a lumbar epidural bolus of lipid soluble opioids is unpredictable. Gourlay et al. (25) demonstrated that peak
cervical CSF concentrations of fentanyl occurred as early as 10 minutes after lumbar epidural administration and averaged 10% of the peak lumbar CSF concentrations. However, in two of the six patients, peak cervical CSF concentrations were twice the level found in the rest of the patients. In another study, sufentanil concentrations were measured after 72 hours of a continuous infusion of 14 μg/h via a low thoracic epidural catheter (26). Sufentanil concentrations in plasma and the cisterna magna CSF were 56% and 82% respectively the concentrations measured in the lumbar CSF. Lipophilic opioids also exhibit rostral migration, but in a less predictable manner than morphine. This difference suggests that the same level of care used for monitoring respiratory depression after epidural administration of morphine should also be utilized when lipophilic opioids are administered for postoperative epidural analgesia. The American Society of Anesthesiologists Task Force has published specific guidelines for the prevention, detection, and management of respiratory depression associated with neuraxial opioid administration (27).

The octanol/buffer partition coefficients and meningeal permeability coefficients for several opioids are presented in Table 13-1.

Morphine was the first opioid approved by the United States Food and Drug Administration (US FDA) for intraspinal administration. Morphine is highly ionized and hydrophilic and does not penetrate lipid-rich tissues as rapidly as fentanyl. Morphine remains within the CSF for a prolonged period of time, spreading rostrally and reaching the trigeminal nerve distribution as early as 3 hours after intrathecal injection in healthy volunteers (24). Morphine requires 45 to 60 minutes to achieve a peak effect, and the duration of analgesia is 14 to 36 hours. This duration may be dose dependent. A variety of intrathecal morphine doses have been investigated. No clear dose–response relationship has been demonstrated using doses greater than 100 μg. Palmer et al. studied patients receiving intrathecal doses from 25 to 500 μg and found a ceiling effect with doses greater than 75 μg, as measured by patient-controlled intravenous morphine use (34). Higher doses conferred no additional analgesic benefit but caused a dose-dependent increase in side effects, particularly pruritus. Palmer et al. also noted that despite high doses of intrathecal
morphine, most parturients continued to administer additional opioid analgesia at a low but constant rate, possibly explaining the interaction between spinal and supraspinal sites of action.

Yang et al. administered 100 or 250 μg of morphine as a component of spinal anesthesia to 60 women undergoing elective cesarean delivery. Women also received 20 μg of spinal fentanyl and perioperative and postoperative nonsteroidal anti-inflammatory drugs (NSAIDs) routinely. There was no significant difference between the small- and large-dose morphine groups in pain relief, as measured by visual analog pain scores (35). Adverse effects of intrathecal morphine have been reported widely and include pruritus, nausea and vomiting, urinary retention, and early or delayed respiratory depression. The most common side effect, pruritus, increased in severity as morphine doses increased. If a morphine dose of 100 μg is used, it is estimated that 43% of women will experience pruritus, and 12% will experience nausea and vomiting (33). Neuraxial morphine administration has also been linked to reactivation of oral herpes simplex. In a study of women with a past history of oral herpes simplex, reactivation occurred in 38% receiving intrathecal morphine compared with 16% of those receiving intravenous morphine (36).

Respiratory depression is an uncommon but potentially serious side effect, though the incidence in the obstetric population is difficult to determine. Abouleish et al. studied 856 parturients who received 200 μg of intrathecal morphine during cesarean delivery and found respiratory depression, as defined by SpO₂ <85% or respiratory rate of less than 10 breaths per minute, in eight patients (0.93%), all of whom were obese (37). The physiologic changes of pregnancy, specifically the higher respiratory rate associated with elevated progesterone levels, may provide a greater margin of safety in comparison to other patient populations. It should be noted however that smaller dosages of intrathecal morphine (75 μg) therapy may result in a reduced duration of analgesia and may therefore require an increased need for supplemental analgesics. Also, due to the variability in patient response to intrathecal morphine, some patients may additionally experience inadequate postoperative analgesia and/or opioid-related side effects.

Fentanyl is arguably one of the most commonly administered intrathecal opioids worldwide. Its relative high lipid solubility results in a greater restriction of segmental activity and rapid onset of action when compared to morphine. Although intrathecal fentanyl offers a relatively short duration of analgesia, it is shown to improve intraoperative analgesia, especially during uterine exteriorization, and provides the patient with a better postoperative transition to other pain medications during recovery from spinal anesthesia. Shende et al. performed a randomized study where 40 healthy patients scheduled for elective cesarean delivery were randomly allocated to receive either saline or 15 μg of fentanyl added to 2.5 mL of hyperbaric bupivacaine intrathecally. They found that the fentanyl group had significantly improved intraoperative analgesia and a longer block regression time. (38). Chu et al. looked at 75 women undergoing elective cesarean delivery and randomized them to receive intrathecally bupivacaine 5 mg alone, or bupivacaine with 7.5, 10, 12.5 or 15 μg of fentanyl. They found that with increasing doses of fentanyl, surgical analgesia was improved and postoperative analgesia lasted longer. They concluded that a ceiling effect was reached at a dose of 12.5 μg. (39). Intrathecal fentanyl also may offer a longer-term analgesic benefit when administered alongside a local anesthetic at the time of cesarean delivery (40). In contrast to morphine, intrathecal fentanyl does not appear to predispose the patient to nausea and vomiting following cesarean delivery. Fentanyl has been shown to cause pruritus in a dose-related manner, though less severely than morphine (32). The risk of delayed respiratory depression is relatively small with intrathecal fentanyl given its segmental effect and lack of rostral spread. A recent large-scale meta-analysis resulted in no reported cases of respiratory depression associated with intrathecal fentanyl at cesarean delivery (33). If respiratory depression does occur however, it usually tends to manifest within the first 30 minutes.

Sufentanil is a thienyl derivative of fentanyl but has higher potency due to greater lipid solubility. The octanol:water partition coefficient of sufentanil is 1,778 and is 91% protein bound. Fentanyl has an octanol:water partition coefficient 813 with protein binding 84%, illustrating the differences in pharmacokinetics. Intrathecal sufentanil offers some theoretical advantages over fentanyl including faster onset, reduced rostral spread, and a lower level of placental transfer. Several studies have been performed comparing fentanyl and sufentanil for analgesia for caesarean delivery and have found them to be equivalent but those women in the sufentanil groups experienced more pruritus (41,42,43). The optimal dose of sufentanil in the subarachnoid space is less than 5 μg with side effects (particularly pruritus) occurring in a dose-dependent manner (44).

Other less-commonly used intrathecal opioids include meperidine, diamorphine, buprenorphine, and, nalbuphine. Meperidine is the only member of the opioid family with local anesthetic-like effects and it has a tendency to sometimes result in motor block. Meperidine has historically been used as a sole spinal drug for cesarean delivery (45).

Diamorphine (3,6-diacetylmorphine) also known as morphine diacetate is a semisynthetic opioid produced by acetylation of morphine. The administration of neuraxial diamorphine for the treatment of pain relief after cesarean section is a common practice in the United Kingdom (46). In contrast, in the United States diamorphine is unavailable for clinical use. Diamorphine has many of the ideal physicochemical properties to provide good pain relief after surgery with the potential to decrease side effects. The intermediate lipid solubility of diamorphine (oil/water partition coefficient 280), increases permeability to both hydrophobic and hydrophilic tissue compartments when compared either with morphine or fentanyl. Diamorphine undergoes metabolism within spinal cord tissue, generating active compounds (6-acetylmorphine and morphine) which increases the analgesic effects, these metabolites are less lipid soluble than the parent drug, that limit their back diffusion into the CSF. Other important physicochemical characteristics of diamorphine are lower PKa (PKa 7.8), low protein binding (40%), and a high unionized fraction (27%) that increases the bioavailability for opioid receptors within the spinal cord and increases clearance from CSF decreasing the potential for serious side effects, such as respiratory depression (47). There is a wealth of data on the use and safety of diamorphine for cesarean delivery in the literature. Cowan et al. looked at 74 parturients for elective cesarean delivery who were randomized to receive either 20 μg fentanyl or 300 μg diamorphine intrathecally using hyperbaric bupivacaine (48). The authors looked at supplemental intraoperative analgesia requirements and found no difference between the groups. When they looked at postoperative
analgesia requirements they found that the diamorphine group had reduced visual analog scale (VAS) scores 12 hours postoperatively but fentanyl reduced VAS only 1 hour postoperatively. They found no difference in pruritus postoperatively between the opioid groups.

There have been three well-reported, dose-finding studies looking at intrathecal diamorphine doses for cesarean delivery (49,50,51). Skilton et al. and Kelly et al. looked at doses up to 0.375 mg and found improved analgesia (as determined by the amount of rescue analgesia required) as the dose increased without a ceiling effect. (49,50). Stacey et al. randomly allocated 40 women undergoing elective caesarean delivery to receive either 0.5 mg or 1 mg of intrathecal diamorphine intrathecally. They found the time to rescue analgesia and 24-hour morphine consumption was significantly lower in the 1 mg group (45% using no opioids postoperatively at all) and that pain scores in this group tended to be lower. Minor side effects were found to be present in both groups but incidence did not differ between groups (51).

Palmer et al. performed a dose–response study looking at 0 to 5 mg epidural morphine on postcesarean pain and found a ceiling effect in terms of analgesia. They found no difference in cumulative systemic morphine use above 3.75 mg used epidurally. 3 mg of morphine epidurally appeared equivalent to 100 μg intrathecally and was found to provide analgesia for 12 to 24 hours (52). There are several studies in the literature comparing epidural and intrathecal morphine for postcesarean analgesia. Sarvela et al. performed a double-blinded, RCT comparing 3 mg epidural morphine with either 100 μg or 200 μg intrathecal morphine and found no significant differences in pain scores. However, rescue analgesia was requested more frequently in the 100 μg group suggesting this dose was less effective for postcesarean analgesia and perhaps unsurprisingly this group were shown to have less pruritus (53).

Due to its prolonged analgesic effects epidural morphine can be administered as an intermittent bolus or as a continuous infusion. It appears that some clinical advantages in using continuous epidural morphine infusions over intermittent bolus for epidural analgesia exist. Studies in the non-obstetric population evaluating morphine’s cephalad migration after a lumbar epidural bolus suggest that respiratory depression may occur as a result of significant amount of the drug reaching the respiratory center in the brain stem after the administration of a bolus dose in the lumbar epidural area (22,23). In fact, large-scale studies suggest that respiratory depression requiring treatment may be higher with intermittent bolus than with continuous infusions (54,55). When mean doses between 7 to 13 mg/d were utilized in the intermittent bolus group and mean doses of 6 to 14 mg/d were used in the continuous infusion group, the incidence of respiratory depression was different. In the bolus study group, the incidence of respiratory depression was 1:500 (54). In contrast, in the continuous infusion group, the incidence was 1:1,500 (55). Based on these data, in the non-obstetric population the maximum risk of respiratory depression within the 95% confidence intervals are 1:100 versus 1:5,000 respectively. Moreover, the concurrent use of parenteral opioids for breakthrough pain, a practice which has been discouraged when intermittent dosing of epidural morphine is used (23), may be administered without an increased risk of delayed respiratory depression even on the surgical wards (54).