Neuraxial administration of morphine includes the epidural or the intrathecal route depending mainly on the type of anesthesia performed (spinal or epidural). There seems to be a better clinical profile for epidural morphine as opposed to its intrathecal administration but there is no clear evidence to recommend one technique over the other. However, the dose under investigation chosen by all studies that have compared the two different routes of administration was based on the doses commonly used in clinical practice rather than the exact potency ratio of intrathecal versus epidural morphine as this remains undetermined. The results therefore, might not be completely reliable.

After epidural administration, variable quantities (depending on which opioid is used) will diffuse across the dura and arachnoid mater into the subarachnoid space to bind opioid receptors in the dorsal horn of the spinal cord. Lipid solubility is the most important factor affecting the rate of diffusion and the subsequent onset and duration of analgesia. Lipophilic opioids such as fentanyl and sufentanil diffuse rapidly across the dura into the CSF compared to hydrophilic opioids such as morphine. Lipophilic opioids produce rapid onset of analgesia which is of short overall duration. After epidural delivery, CSF opioid levels peak at 6 min for sufentanil, 20 min for fentanyl, and 1–4 h for morphine. The epidural space is extremely vascular and there is extensive absorption of opioids via the epidural venous plexus into the systemic circulation. Systemic opioids reach the CNS and bind receptors in areas of the brain that modulate pain perception and response.

Morphine is currently the “gold-standard” neuraxial opioid for post-cesarean analgesia primarily due to its long-acting effect that can last for many hours after a single administration [16].

Morphine is a hydrophilic agent that is not easily absorbed by the blood vessels and fat and therefore it has high central nervous system availability and a very long duration of action.

Epidural morphine is generally administered as a single bolus rather than with a continuous infusion technique.

The administration of epidural morphine follows a precise dose–response relationship: as the dose of epidural morphine is increased, the quality and duration of analgesia increases accordingly until a sailing effect is reached, at 3.75 mg, where the quality of analgesia does not increase [17].

Many studies have been performed in order to find the best balance between drug efficacy and reduced side effects for epidural morphine. The administration of 2–4 mg of epidural morphine is suggested as it can provide optimal analgesia for the first postoperative day, minimizing at the same time the risk of side effects such as nausea, vomiting, and respiratory depression [17–20].

Side effects after epidural morphine administered as a single 2–4 mg bolus are characterized by mild pruritus that can occur in approximately 50% of patients, nausea and vomiting in approximately 30–40% of patients, and dizziness in 10%.

Respiratory depression after epidural morphine may occur early after its administration, approximately in the first hour thanks to its vascular absorption or it can be observed later, after approximately 6–18 h due to its slow cephalad spread in the cerebrospinal fluid and consequent penetration into the brainstem [21, 22]. The incidence of respiratory depression after clinical doses of epidural morphine is unlikely to occur. However, if the doses are increased above 5 mg, even young and healthy parturients can develop clinically detectable respiratory depression. In such cases, clinical respiratory monitoring might be indicated.

Both fentanyl and sufentanil are highly lipophilic opioids that act with different affinity for the mu receptor [23]. They are both characterized by a fast onset of analgesia and shorter duration when compared to morphine due to their high lipid solubility. Epidural fentanyl 50–100 μg or sufentanil 10–20 μg represent the dose currently used in clinical practice and usually provide effective analgesia that lasts for approximately 4–5 h. The increase in the dose of these opioids is not associated with an increase in the efficacy or the duration of action.

When used epidurally, sufentanil is 5.9 times more potent than fentanyl [24] although no differences were observed in both the onset and the duration of analgesia between the two opioids [25].

Neuraxially administered fentanyl and sufentanil can be effectively used to improve intraoperative analgesia and provide early postoperative analgesia [26] while their short duration make them unsuitable agents for the treatment of post-cesarean analgesia.

The side effects associated with these two short-acting opioids are comparable to that observed with the administration of morphine but with an earlier onset.

Both these two opioids can cause early onset respiratory depression in less than 1 h after their administration due to vascular absorption and rostral spread in cerebrospinal fluid [27].

Spinal anesthesia is the most commonly used technique for cesarean delivery; therefore, the addition of opioids to the spinal solution in order to enhance and prolong intraoperative and postoperative analgesia has become the standard practice worldwide.

Intrathecal opioids exert their action primarily by directly binding pre- and postsynaptic mu-opioid receptors in the substantia gelatinosa of the dorsal horn of the spinal cord and they are also transported supra-spinally by CSF flow where they modulate descending inhibitory pain pathways.

A large variety of opioids have been investigated as suitable options for postoperative analgesia such as morphine [28–30], fentanyl [30, 31], and sufentanil [32].

Morphine is the most commonly used intrathecal opioid as it can provide excellent analgesia with a long duration of action. Analgesia provided by morphine is generally characterized by a slow onset and a long duration of action, generally up to 24 h [33].

Unlike epidural morphine, its intrathecal administration does not follow a precise dose–response relationship for analgesia. Dose–response studies have in fact found that the analgesic efficacy of intrathecal morphine increases until the dose of 50–150 μg [28]. The incidence of side effects such as nausea and vomiting did not appear to follow a dose-related effect while the incidence of pruritus increases in a dose-dependent fashion and can be observed in up to 90% of the cases [28, 33].

The optimal intrathecal dose for morphine was investigated by several studies and meta-analyses which examined doses ranging from 0.1 to 0.5 mg and found that doses from 0.1 to 0.2 mg are associated with optimal analgesia lasting up to 27 h (from 11 to 29 h) and reduced side effects while doses above 0.2 mg do not provide an improvement in the quality of analgesia [20]. However, the use of 0.2 mg intrathecal morphine instead of 0.1 mg is associated with only a little improvement in the analgesic efficacy, but with a twofold increase of side effects such as nausea, vomiting, use of antiemetics, and pruritus. The use of 0.1 mg, therefore, might represent the preferable choice for nursing parturients although in most cases it is still associated with nausea and pruritus [34].

One other potential complication that can be observed with intrathecal morphine is the rostral spread in the cerebrospinal fluid and consequent penetration in the brainstem due to its extremely low lipid solubility that can possibly lead to late respiratory depression. This however, is much more frequent when administered intravenously, and with the intrathecal doses commonly used in clinical practice it is very unlikely to observe this harmful complication.

Nevertheless, it is important to take into account that the sole administration of morphine either intrathecally or epidurally is often accompanied by the request for additional pain medications (multimodal approach).

A randomized controlled trial examined the dose–response relationship of intrathecal morphine comparing 0.05, 0.1, or 0.15 mg combined with 30 mg intravenous ketorolac in patients undergoing elective cesarean section with 12 mg of hyperbaric bupivacaine and fentanyl 15 μg. The results of the study indicate that 0.05 mg of intrathecal morphine produces similar analgesia as 0.1 or 0.15 mg when used as part of a multimodal therapy. The only difference observed was the incidence of pruritus, greater in the 0.1 and 0.15 mg, while the incidence of nausea and vomiting was comparable in the three groups [35].

Intrathecal fentanyl and sufentanil can be used to provide optimal intraoperative analgesia as they are highly lipid soluble so characterized by a fast onset of action. In fact, when added to intrathecal bupivacaine they allow for a reduction in antiemetic medications and provide early postoperative analgesia. However, they also have a brief duration of action and are therefore not suitable as postoperative agents [32].

The use of intrathecal fentanyl and sufentanyl were found to provide some early postoperative analgesia [30] and contributed to an increase in the quality of intraoperative analgesia in a small percentage of parturients [20, 32].

Diamorphine is a lipophilic opioid that rapidly diffuses in the cerebrospinal fluid after its administration [36] and can be used either epidurally or intrathecally. It is available only in the UK.

In comparison to morphine, diamorphine is more lipid soluble so its onset of action is shorter while its duration is comparable to that obtained with morphine into which it is rapidly metabolized once it reaches the CSF. In terms of safety, diamorphine is unlikely to produce late respiratory depression as there is little drug that gains access to the spinal cord and brainstem from the cerebrospinal fluid due to its rapid clearance.

Its efficacy after being administered epidurally and intrathecally has been investigated in several trials.

Dose–response studies have found that increasing the dose of intrathecal diamorphine is associated with an increase in analgesia efficacy without a ceiling effect and with a concurrent increase in the incidence of side effects [37, 38].

One study which investigated the ED95 of intrathecal diamorphine suggested the use of 0.4 mg in clinical practice that was able to provide effective analgesia with a mean duration of action of approximately 10 h [39]. However, this was obtained at the expense of a significant increase in the incidence of side effects such as nausea and vomiting and pruritus observed in more than 50% of parturients. For this reason, many authors have suggested reducing the dose of this analgesia medication to 0.3 mg [39, 40].

Epidural diamorphine has a fast onset and a long duration of action and it is clinically used in doses ranging from 2.5 to 5 mg with good analgesic efficacy and a duration of approximately 14 h [41, 42]. The increase in the dose of the drug is associated with an increase in the duration of analgesia at the expense of an increase in the incidence and severity of side effects. For this reason, the suggested dose for epidural diamorphine is about 3 mg [43].

The epidural administration of diamorphine was found to be as equally effective as its intrathecal administration with regard to both prolonged duration and good quality of analgesia [42], although the epidural administration is associated with an increased incidence of side effects such as nausea and vomiting. The authors of the study in fact suggest the use of intrathecal diamorphine due to its favorable side effect profile as part of a spinal or a combined spinal epidural technique.

Recently extended-release epidural morphine (EREM) formulations have been investigated in comparison to the conventional neuraxial morphine that can provide good post-cesarean analgesia but for only 1 day after surgery.

Extended-release epidural morphine was found to significantly improve pain scores after cesarean section and prolong post-cesarean analgesia when compared with conventional epidural morphine, with no significant increase in the incidence and severity of side effects [43, 44].

The administration of systemic morphine is commonly used for postoperative analgesia when cesarean section is performed under general anesthesia and consists of the intravenous, intramuscular, and subcutaneous routes.

The intramuscular and subcutaneous routes are less commonly used when compared with the intravenous route as the use of repeated injections is uncomfortable for the parturients, and because there is a large interindividual variability in opioid requirement, onset of action, duration, and pharmacokinetics [14], although they are associated with fewer side effects. These latter factors can contribute to intermittent and suboptimal levels of analgesia obtained with intramuscular or subcutaneous administrations.

Morphine is often administered intravenously as part of a patient-controlled intravenous anesthesia technique (PCIA). Its advantages are represented by more stable levels of analgesia due to the low fluctuations in plasma opioid levels and great analgesic efficacy when compared with the intramuscular administration that leads to higher maternal satisfaction [45]. In addition, the feeling of control that women can have with using PCIA administration contributes to providing good overall parturient satisfaction.

On the other hand, major disadvantages of the PCIA technique are that women must be correctly instructed on the proper use of the device and also that new mothers are often concerned about the potential entry of the drug into their milk leading to a reduction in the demand doses.

Some authors suggest the use of the PCIA technique associated with a background infusion of the solution. However, the efficacy of this technique is controversial [46], more side effects can be seen, and also there is concern on its safety.