Evaluation of Therapies for Peripheral and Neuraxial Opioid-induced Pruritus based on Molecular and Cellular Discoveries

Although neuraxial opioids are frequently used in acute perioperative pain, they also have side effects. These side effects include pruritus,^10–13  nausea (25%), sedation (17%), urinary retention (19%), and, very rarely, respiratory depression (3%).^9,16,17  Although pruritus is occasionally observed with parenteral opioid use,^18,19  it is an extremely common side effect of neuraxial opioids, with an incidence ranging from 30 to 85%, depending on the dose and lipophilicity of the opioid administered.^12,20,21  Dose-response relationships between spinal morphine, analgesic duration, and pruritus have indicated that escalating doses of morphine improve analgesia but are correlated with higher incidences and severity of pruritus.^22,23  This relationship suggests that the duration of analgesia of neuraxial morphine can be improved with increased doses but xst be weighed against increasing side effects.

Because of the widespread use of neuraxial opioids in the childbirth setting, opioid-induced pruritus is most frequently observed among obstetric patients^16,24–26  (table 1), where it has an incidence of up to 85%,²⁰  and is primarily dose-dependent.^22,23  Pruritus is also frequently reported among orthopedic patients receiving neuraxial opioids (30 to 70%)^37,38  for Enhanced Recovery after Surgery. The incidence of pruritus in patients receiving neuraxial morphine is more common among female patients (60 to 85%),^16,24–26  which mirrors the incidence and burden of chronic pain that also disproportionately affects women.³²  Differences in estrous cycle, which could affect the sensitivity of the μ-opioid receptor to opioid drugs,³⁹  have been proposed to underlie these differences, yet the extent to which nonbiologic factors could also contribute to disparities in experiences of pruritus warrants further investigation.

Although it is less common than neuraxial opioid–induced pruritus, parenteral opioid–induced pruritus has been observed in 10 to 50% of patients, particularly in the setting of patient-controlled analgesia (PCA)^15,27,40,41  and among patients receiving IV morphine for vasoocclusive crisis in sickle cell disease.⁴²  Thus, in contrast to neuraxial opioid–induced pruritus, which is observed after a single dose of opioids, parenteral opioid–induced pruritus is mostly observed after extended dosing.^15,27,40,41  Among pediatric patients on PCA, pruritus was the most common reason for switching medications to another opioid class: for example, from morphine to a semisynthetic opioid such as hydromorphone.²⁷  In these settings, the development of pruritus may be associated with central sensitization to opioids that may occur independently of tolerance to opioid-induced analgesia.⁴³  Proposed mechanisms include sensitization of itch-responsive circuits within the spinal dorsal horn after chronic use.⁴³  In particular, central, but not peripheral, μ-opioid receptors have recently been implicated in modulation of dermatitis and lymphoma-induced chronic itch.⁴⁴  Further research is necessary to elucidate the contribution of these central pathways to chronic opioid-induced pruritus. Thus, although pruritus is most common after neuraxial opioid administration, it also poses a troublesome side effect for a small number of patients, such as those on PCA for postoperative pain management.

Sequencing of the human μ-opioid receptor (OPRM1) gene, which encodes for the μ-opioid receptor, indicates that certain polymorphisms may be associated with protection against the side effect of pruritus.^28,46  For example, genetic association studies have focused on the A118G polymorphism of OPRM1.^28,46  The recessive G allele in this polymorphism is associated with lower incidences of pruritus among obstetric patients receiving epidural (4.8%) and spinal (0 to 50%) morphine.^28,46  This variant is further associated with reduced sensitivity to the analgesic effects of opioids,⁴⁷  suggesting that the A118G polymorphism may give rise to a μ-opioid receptor that is generally less responsive to opioid medications. However, given the heterogeneity among clinical populations receiving opioid analgesia and their self-reported experiences for both pain and itch, many of these studies are underpowered to provide definitive conclusions about these polymorphism relationships.^48,49  In mice, distinct splice variants of the u-opioid receptor isoform consisting of exons 1-3 and 8-9 (MOR1D) and exons 1-4 (MOR1), have been shown to differentially modulate morphine-induced itch and morphine-induced analgesia, respectively.⁵⁰  However, the contributions of these splice variants have not been characterized in humans. Additional large population genetic association studies are necessary to further assess the clinical significance and utility of identifying genetic variations on acute postoperative pain management and risk for opioid-related side effects.

Opioid-induced pruritus is observed after the use of intrathecally administered hydrophilic opioids (such as morphine) and lipophilic opioids (such as fentanyl and sufentanil).^29,51  The high incidence of pruritus after morphine (60 to 85%)^{10,11,13,19,30}  and lipophilic opioids (60 to 90%)^29,51  suggests a common μ-sensitive pathway modulates pruritus in response to these medications (table 1). However, a key difference between these two classes of opioid drugs is the onset and duration of pruritus; pruritus occurs within hours and lasts up to several days in patients receiving epidural and spinal morphine.^30,31  With lipid-soluble opioids, the onset of pruritus can occur as rapidly as within minutes of administration and persists up to several hours.^25,29  These differences in onset and duration reflect the pharmacokinetics of these intrathecally administered opioid medications. Microdialysis studies in pigs³²  (using equimolar doses of morphine, fentanyl, and sufentanil) have revealed that spinal exposure to morphine was greater than lipophilic opioids because of morphine’s low spinal cord distribution volume and slow clearance into the plasma. Fentanyl and sufentanil, in contrast, have been found to rapidly clear into the plasma and epidural fat, respectively, reducing their spinal bioavailability.³²  In a clinical study examining the clearance of opioid medications from the cerebrospinal fluid (CSF), volunteers received an intrathecal injection of both morphine and fentanyl (50 μg each). The ratio of morphine to fentanyl in the CSF was found to increase over time (reaching 4:1 within 2 h).⁵²  The solubility of morphine compared to lipid-soluble opioids, such as fentanyl, clearly affect the bioavailability and thus onset and duration of opioid analgesia and its side effects. Furthermore, the bioavailability of neuraxial morphine supports its suitability for acute postoperative pain, such as for Enhanced Recovery after Surgery⁵³  and in obstetric settings⁵⁴ ; however, its long elimination time must be considered against its delayed adverse effects, such as pruritus.

It has been proposed that release of histamine underlies opioid-induced pruritus,⁵⁵  although the route of administration might lead to differential impact of mast cells on pruritus (fig. 1). Intramuscular and subcutaneous morphine have been shown to evoke pruritus and vasodilation at the injection site.⁵⁶  One study that performed in vivo microdialysis in human skin found that intradermal injections of morphine led to dose-dependent increases in local histamine and itch sensations.⁵⁷  Similarly, examination of blood samples from patients with parenteral exposure to opioids has also revealed an elevation in histamine levels.^33–35  Last, in vitro studies in both human and rodent models have confirmed that opioids, such as morphine, act directly on mast cells to drive degranulation and histamine release.^58–60 

Only recently, however, has it become clear how opioids could directly activate mast cells. Morphine was recently found to induce mast cell degranulation through activation of mas-related G protein-coupled receptor X2 (MRGPRX2), a primate-exclusive G protein–coupled receptor, rather than canonical opioid receptors (such as the μ-opioid receptor).⁵⁸  Binding of opioids to MRGPRX2 leads to increases in intracellular calcium in mast cells.⁵⁸  This is thought to occur through the phospholipase Cβ pathway, which results in the release of mediators, such as histamine.⁶¹  MRGPRX2 is implicated in itch and pain,⁶²  and the restricted expression of MRGPRX2 on mast cells, which can be activated in the absence of atopy,⁶¹  such as by direct binding of opioids,⁵⁸  establishes a mechanism for how opioids can directly act on mast cells to give rise to pruritus (fig. 2). The expression of MRGPRX2 on mast cells may also underlie anaphylaxis after opioid administration, which comprised 2.6% of all cases of anaphylaxis caused by anesthetics.⁶³  For these reasons, MRGPRX2 should be considered a target for IgE-independent allergic reactions to perioperative drugs such as morphine administered subcutaneously and intravenously.

However, for neuraxial opioids, the evidence that a mast cell–dependent mechanism underlies pruritus is less compelling. In humans, the neuraxial administration of fentanyl, a lipophilic and synthetic opioid, evokes pruritus, even though fentanyl does not cause mast cell degranulation and histamine release as morphine does.^60,64–66  Furthermore, clinical plasma concentrations of morphine are orders of magnitude smaller than CSF concentrations after neuraxial morphine administration (less than 0.01%),^67,68  and the concentration of morphine detected in the plasma is likely insufficient to cause mast cell degranulation, as suggested by in vitro studies (fig. 2).^58–60  Mast cell–mediated pruritus after the local injection of opioids, as seen after intradermal injections of morphine, is restricted to the site of injection⁵⁷  and does not explain how neuraxial opioids, frequently administered into lumbar segments, evoke pruritus in other dermatomes, including those innervating facial regions.^10,11,13  Last, genetically modified mice lacking mast cells scratch at levels similar to controls in response to intrathecal morphine.³⁶  Therefore, a mast cell-dependent mechanism does not explain the prevalence and spread of neuraxial opioid–induced pruritus. Taken together, although there is evidence that histamine release may cause itch from peripheral (subcutaneous, oral, and intravenous) opioids, this process is unlikely to contribute to neuraxial opioid–induced pruritus.

With neuraxial morphine-induced itch, it has been proposed that the nervous system could be responsible for mediating the sensation of itch^12,69  (fig. 2). The analgesic benefits of opioids are mediated through the nervous system^70,71 ; therefore, it is proposed that pruritus, another sensory symptom, may also be mediated through neurons. Although both intrathecal and epidural opioids cause pruritus,^10,11  pruritus occurs more commonly with the intrathecal route of administration; a review of 11 trials by Simmons et al.⁷²  found the average relative risk of developing pruritus in combined spinal analgesia relative to epidural analgesia to be 1.80 (95% CI, 1.22 to 2.65).⁷²  This supports the possibility that pruritus occurs because of a direct effect of the opioid within the neuraxis.

Further evidence for a neuronal mechanism for pruritus is the observation that opioid lipophilicity is associated with the onset and severity of pruritus among patients. For example, lipophilic opioids, such as fentanyl, which more readily cross the blood–brain barrier, are associated with a more rapid onset and shorter duration of pruritus than hydrophobic opioids such as morphine,^12,21,25,73  consistent with a neuronal mechanism of action. These clinical observations further support the idea that two separate mechanisms underlie opioid-induced pruritus arising from different routes of administration (fig. 2): systemic opioids activate mast cells to drive itch through histamine release, whereas neuraxial opioids act on CNS pathways to cause itch.

Primary sensory neurons respond directly to itch-inducing agents. For example, a class of transient receptor potential cation channel subfamily V member 1 (TRPV1)-containing C-fibers that innervate the skin expresses the histamine receptor, which can be activated by histamine.^74,75  Upon activation with histamine, these neurons transmit pruriceptive information to the superficial layers of the spinal cord, and this information ascends through the spinothalamic tracts to target the primary sensory and cingulate cortices.⁷⁶  Thus, if opioid-induced pruritus causes the release of histamine, then peripheral opioid-induced itch may occur through this pathway (fig. 1). Consistent with this view, intradermal injection of the μ-opioid receptor agonist D-Ala2, N-MePhe4, Gly-ol]-enkephalin (DAMGO) elicits itch in mice that is abrogated with the ablation of TRPV1-expressing fibers.⁵⁶  Another channel expressed by primary afferents is transient receptor potential cation channel, subfamily A, member 1 (TRPA1), which has also been implicated in acute and chronic itch.^77,78  However, the contribution of TRPA1 to peripheral opioid-induced itch is unlikely because TRPA1-deficient mice exhibit normal histamine-evoked itch.^77,78  These findings suggest that for subcutaneous, oral, and intravenous opioids, itch likely occurs through TRPV1, and not TRPA1, sensory neurons after the release of histamine from mast cells.^33–35  However, given the weak association between neuraxial opioids and histamine release,^60,64–66  histamine receptor–expressing sensory neurons are not likely to be involved in neuraxial opioid–induced pruritus.

Presently, neuraxial opioids are not believed to act directly on μ-opioid receptor expressing sensory neurons to elicit itch. Conditional deletion of the μ-opioid receptor from TRPV1 and somatostatin neurons, which have been implicated in itch signaling, did not affect intrathecal morphine–induced itch in mice.⁴⁵  In support of this view, other studies have suggested that rather than direct modulation of these peripheral sensory neurons, neuraxial opioids may influence pruriceptive processing centrally. One study, performed in rats, identified that neuraxial morphine augmented the activity in trigeminothalamic tract neurons in response to itch stimuli.⁷⁹  Trigeminothalamic tract neurons are analogous to spinothalamic tract neurons in the spinal cord but relay information pertaining to the head and neck regions rather than the body. It was found that intrathecal morphine caused these neurons to increase their ongoing activity to pruritogens applied to the skin.⁷⁹  Intrathecal morphine also enhanced responses to innocuous mechanical stimuli, suggesting that opioids may also cause sensitization to touch-evoked itch.⁷⁹  Another study in mice also examined the role of TRPV1 antagonists and neuraxial morphine-induced itch.⁸⁰  In this study, it was observed that the intrathecal delivery of a TRPV1 antagonist reduced morphine-induced itch in mice. Together, these studies suggest that neuraxial morphine can evoke sensitization of central pathways, leading to an enhancement of responsiveness within these circuits to peripheral stimulation.

It is curious that opioids suppress pain but evoke itch. Conversely, intense pain can also suppress itch.^81,82  Previously, it was believed that itch occurs because of a reduction in pain signaling.⁴⁴  However, detailed pharmacologic investigations have largely ruled out this theory.^50,83  Several studies have shown that μ-opioid agonists, such as morphine, elicit itch, but δ- and κ-opioid receptor agonists, which also produce analgesia,⁸⁴  do not cause scratching behavior.^69,83,85  Instead, selective κ-opioid receptor agonists have been shown to suppress itch in preclinical models^86–88  and have been approved to treat chronic pruritus in Japan.⁸⁹  Notably, μ and κ-opioid receptors often exert opposing effects in several regions of the central nervous system; whereas μ-opioid receptor agonists produce analgesia, euphoria, and itch, κ-opioid receptor agonists are antipruritic and dysphoric.^90,91  Clearly, a simple model whereby a suppression of pain is sufficient to evoke itch does not explain these observations. The differential roles of these opioid receptors and their agonism in the context of itch underscores that opioid-induced pruritus is a complex and active process.

The mechanisms of opioid-induced itch at the level of the dorsal horn have only recently begun to be examined. One study in mice suggested that neuraxial morphine causes itch through activation of gastrin-releasing peptide receptor neurons in the dorsal horn,⁵⁰  an excitatory population that is involved in mediating itch (fig. 3).^92,93  This study suggested that neuraxial morphine triggers heterodimerization of the μ-opioid receptor with gastrin-releasing peptide receptor.⁵⁰  After heterodimerization, the authors proposed that itch occurs after the activation of excitatory downstream pathways involving phospholipase Cβ3 and an increase in intracellular calcium.⁵⁰  However, recent sequencing^94,95  and neurochemical studies^36,45  cast doubt on this conclusion. Opioids typically signal through Gαi-coupled G protein–coupled receptors, resulting in potassium efflux and hyperpolarization, which inhibits neuronal activity.^96,97  Thus, the conclusion that neuraxial opioid–induced itch occurs through dimerization of the μ-opioid receptor and gastrin-releasing peptide receptor is controversial.⁶⁹ 

Alternatively, molecularly defined inhibitory neurons have been implicated in the modulation of itch. For example, inhibitory neurons containing dynorphin and neuropeptide Y have been shown to be important for the inhibition of chemical^88,98,99  and mechanical itch,^100,101  respectively. When either dynorphin neurons or neuropeptide Y neurons are lost during development, mice show spontaneous scratching behavior suggesting that these populations are involved in the tonic inhibition of itch.^88,98–101  In a recent study, selective removal of the μ-opioid receptor from inhibitory neurons (neurons that produce γ-amino butyric acid [GABA]) abolished opioid-induced itch in a mouse model.⁴⁵  A second study focused on a subset of inhibitory neurons containing dynorphin, the endogenous peptide for the κ-opioid receptor.³⁶  In this study, the expression of the μ-opioid receptor was found to be required for morphine-induced itch.³⁶  κ-opioid signaling alleviated morphine-induced itch in both mice and nonhuman primates.³⁶  Therefore, emerging evidence highlights that opioids could cause itch through inhibition of inhibitory neurons; rather than through heterodimerization between the μ-opioid receptor and gastrin-releasing peptide receptor, there is now compelling evidence that opioids cause itch through a mechanism of neuronal disinhibition (fig. 3).

These two recent studies, from two independent groups, highlight the role of inhibitory neurons in the spinal cord as the crucial mediators of opioid-induced itch. To date, they provide the most compelling evidence for how a spinal mechanism is responsible for neuraxial opioid–induced itch.

Given the dose-dependent nature of opioid-induced pruritus, clinicians have managed to control unwanted side effects by tightly titrating the dose needed to optimize analgesia.¹⁰²  Multimodal analgesia has also been shown to be effective at both managing pain and reducing opioid side effects through synergy between multiple agents.¹⁰³  For example, the combination of local anesthetics, such as bupivacaine, with opioid analgesics reduces the severity of pruritus in the immediate postoperative period.^51,73  However, despite efforts to reduce the dose of opioids administered and to apply a multimodal approach to neuraxial analgesia, pruritus as a side effect persists for some people. Common and modern treatment options for these patients are described below (table 2).

The role of antihistamines in neuraxial opioid–induced pruritus is contested. Antihistamines reduce diaphoresis and wheal-and-flare responses to parenteral opioids,³⁴  and many providers continue to use antihistamines to manage pruritus induced by opioids. Nevertheless, the appropriateness of this practice in the setting of neuraxial opioid–induced pruritus is questionable.¹²⁴  Several studies indicate that histamine receptor antagonists, such as diphenhydramine and promethazine, reduce itch in obstetric patients who receive neuraxial opioids.^{104.124–127}  However, antihistamines have also been shown to cause sedation,^19,105,106  and it has been observed that the sedating effects of antihistamines have caused patients to verbally deny itch but continue to scratch or that patients report itchiness in between periods of sleep.^10,104,106  Thus, it is possible that the apparent reduction in itch with antihistamine treatment seen in some studies may, in fact, be secondary to drowsiness.¹²⁸ 

Comparisons between the effectiveness of mixed κ agonist–μ antagonist and antihistamines for the management of neuraxial opioid–induced pruritus have revealed major limitations of antihistamine treatment for this form of pruritus.^{104,125,126,129–131}  In spite of these limitations, they continue to be prescribed for both peripheral and central opioid-induced pruritus.^12,127  Given new evidence that spinal neurons mediate neuraxial opioid–induced pruritus,^45,36,69  antihistamines likely have no role in the treatment of neuraxial opioid–induced pruritus, but they may have a key role in the treatment of subcutaneous, oral, and intravenous morphine–induced pruritus.

To date, the most effective treatments for opioid-induced pruritus have included pharmacologic agents that antagonize the μ-opioid receptor, which pose disadvantages in that they can reverse analgesia.^107,132  Clinically, μ-opioid receptor antagonists have also been shown to be effective for the management of parenteral opioid–induced pruritus.¹²⁹  For patients on parenteral opioid therapy, such as patients with sickle cell disease, the coadministration of the opioid analgesic and its antagonist helps to mitigate pruritus,⁴²  particularly when small doses are infused (e.g., naloxone 0.25 to 2 μg · kg^–1 · h^–1 intravenous push).^133,134  Naloxone is also effective at reducing wheal-and-flare responses caused by morphine.¹³⁵  These findings highlight the utility of μ-opioid receptor antagonists in the management of peripherally mediated and histamine-dependent opioid induced pruritus.

μ-Opioid receptor antagonists are also effective for the treatment of pruritus induced by neuraxial opioid administration.^12,13  Naloxone and naltrexone are direct opioid receptor antagonists often used to reduce both the frequency and the severity of pruritus evoked by neuraxial opioid analgesia.¹³²  Unfortunately, these antagonists, at doses that are clinically effective at reversing pruritus, may also reverse the analgesic effects of opioids⁴¹  (e.g., complete reversal for opioid-induced respiratory depression at doses of naloxone 0.1 mg/kg intravenous push).¹⁰⁸  In nonhuman primates, the one-time administration of a selective μ-opioid receptor antagonist, such as nalmefene, produced a 10-fold rightward shift in both morphine-induced scratching and analgesia.⁸³  Furthermore, the mean pK_B (an estimate of antagonist affinity) of nalmefene was found to be similar for both scratching and thermal nociception endpoints,⁸³  indicative of a circumscribed window for the management of itch without affecting analgesia by μ-opioid receptor antagonism, making these agents suboptimal for the treatment of pruritus in obstetric patients. Thus, optimal doses of μ antagonists, such as naloxone or naltrexone, to relieve clinical pruritus often come with a risk-benefit discussion with patients about their preferential priorities on pain control versus itch.

Alternatively, mixed opioid receptor agonists, such as nalbuphine (1 to 5 mg IV)¹³⁶  and butorphanol (0.2 to 2 mg IV)¹³⁷  are also clinically effective therapies for neuraxial opioid–induced pruritus.^19,107,138  They are frequently used due to their improved ability to manage pruritus without reducing analgesia compared to selective μ-opioid receptor antagonists.¹⁰⁷  Nalbuphine is a mixed antagonist of the μ-opioid receptor and agonist of the κ-opioid receptor, and because of its partial antagonism of the μ-opioid receptor, reversal of analgesia remains a concern.¹⁰⁹  Another potential limitation of nalbuphine is its sedating side effect,^19,133  although this is only observed when a high dose (10 mg/70 kg) is used,¹³⁹  which is beyond the range used to manage pruritus clinically.²⁴  In contrast, butorphanol is a partial agonist of both the μ- and κ-opioid receptors. Similar to nalfurafine, butorphanol has also been reported as an effective treatment of opioid-induced pruritus, particularly among pediatric patients.^140,141  Thus, nalfurafine and butorphanol may pose therapeutic advantages over selective μ-antagonists because of their agonism of the κ-opioid receptor and the ability to directly modulate spinal itch circuits.

Notably, both opioid antagonists and mixed agonist–antagonists are effective for other clinical instances of pruritus, including those associated with systemic disease or of dermatologic origin. Nalbuphine, for example, is effective at managing pruritus among patients with end stage renal disease^142,143  and contact dermatitis.¹⁴⁴  These observations suggest that different forms of pruritus (spanning from systemic disease to drug-induced) converge on a common itch pathway that depend on central opioid signaling, likely centered upon the involvement of endogenous endorphin and dynorphin tone. Like many drugs frequently used in clinical anesthesia practice, the use of nalbuphine for the clinical management of opioid-induced pruritus is off-label.¹⁴⁵  Given the effectiveness of nalbuphine for the prevention and severity of pruritus,¹⁹  a new drug indication of nalbuphine for opioid-induced pruritus could facilitate its widespread use within clinical practice.

The dynorphin–κ-opioid receptor system has been heavily implicated in itch. Pharmacologic and genetic manipulations of spinal dynorphin-expressing neurons in freely behaving animals have uncovered that spinal dynorphin is required for the inhibition of itch.^88,99  In rodent and nonhuman primate models, selective κ-opioid receptor agonists, such as nalfurafine, have been shown to be important for the inhibition of several forms of chemical, immunologic, and drug-induced (including opioid-induced) itch.^87,146,147  Clinically, nalfurafine (2.5 or 5.0 μg per os)¹⁴⁸  has been shown to be effective for the treatment of uremic and cholestatic pruritus in Japan.⁸⁹ 

Emerging evidence in preclinical models, involving both mice and nonhuman primates, indicate that the intrathecal and systemic administration of κ-opioid receptor agonists can reduce morphine-induced itch without reducing morphine-induced analgesia.^36,87,149  In preclinical models, nalfurafine has been found to restore dynorphin signaling disrupted by neuraxial μ-opioid receptor agonists, such as morphine.³⁶  These preclinical findings indicate that selective κ-opioid receptor agonists should be further considered for opioid-induced pruritus in the future.

Presently, however, nalfurafine does not have approval by the European Medicines Agency (Amsterdam, The Netherlands) or Food and Drug Administration (Silver Spring, Maryland) for clinical use in Europe and the United States, respectively.¹⁵⁰  One concern has been its sedating effects, which has been observed after long-term use in dogs,¹⁵¹  but this has not been a consistent finding in other animal models and in patients.^87,148,152  In European trials, nalfurafine (5 μg per os) did not significantly reduce the severity of uremic pruritus compared to placebo over 8 weeks.¹⁵¹  However, one key difference between the studies conducted in Europe¹⁵¹  and Japan¹⁵³  was the duration of the trial (weeks compared to days, respectively), and notably, in the European trials, both the nalfurafine-treated and placebo groups exhibited significant reductions in self-reported visual analogue scale intensities of pruritus over the course of the study.¹⁵¹  Given the incidence of opioid-induced pruritus in the postoperative setting, the effectiveness of nalfurafine may be more evident in acute settings, although further clinical trials in perioperative and acute care settings is necessary.

5-Hydroxytryptamine receptor antagonists are frequently used to treat postoperative nausea and vomiting. Both of these side effects are observed in postoperative patients who receive neuraxial morphine.^154,155  Several studies have revealed that prophylaxis with a 5-hydroxytryptamine receptor antagonist (such as ondansetron, ranging from 4 to 8 mg IV) does not significantly reduce the incidence of pruritus.^{20,110–114,154}  A systematic review by Bonnet et al.¹¹⁵  of 15 randomized controlled trials suggests that 5-hydroxytryptamine receptor antagonists may reduce the intensity of opioid-induced pruritus but also concluded that the trials included in the systematic review, such as small studies that favor the publication of positive findings, may have suffered from publication bias. Thus, there is still a lack of consensus on the clinical effectiveness of 5-hydroxytryptamine receptor antagonists in opioid-induced pruritus.^115,156 

The mechanism by which 5-hydroxytryptamine receptor antagonists such as ondansetron alleviate itch is not clear. 5-Hydroxytryptamine receptor immunoreactivity has been observed in the spinal dorsal horn,¹⁵⁷  and the endogenous source of serotonin is thought to originate from descending fibers arising within the brainstem.¹⁵⁸  Depletion of supraspinal serotonin has been shown to alleviate pruritogen-induced itch in rodents,¹⁵⁹  which may partially explain how antagonism of 5-hydroxytryptamine receptor reduces opioid-induced itch. However, spinal neurons expressing 5-hydroxytryptamine receptor and the μ-opioid receptor have been found to comprise nonoverlapping populations,⁹⁵  making it unlikely that agents such as ondansetron could directly reduce the activity of neurons responsible for opioid-induced itch. In rhesus monkeys, ondansetron has also been found to be ineffective at reducing morphine-induced itch, even at high doses that caused extrapyramidal effects.⁶⁹  Given these conflicting reports between human and animal models and the lack of a cellular basis for how serotonin and opioid signaling could converge, 5-hydroxytryptamine receptor antagonists are currently not considered front-line therapies for opioid-induced pruritus and likely have a limited role in prophylaxis against opioid-induced pruritus.

The effectiveness of subhypnotic doses of propofol (10 to 30 mg IV) for the treatment of opioid-induced pruritus remains controversial. In one double-blind trial, propofol (10 mg IV) was observed to reduce morphine-induced pruritus compared to placebo (85% compared to 16%, respectively).¹⁶⁰  The patients in this study had received either epidural or spinal morphine for a variety of surgical procedures including gynecologic, gastrointestinal, thoracic, and orthopedic surgery.¹⁶⁰  Another study also reported that propofol (10 mg IV) protected against pruritus after intrathecal morphine for arthroplasty surgery.¹¹⁶  However, Warwick et al.¹¹⁷  did not confirm these findings in a double-blind study of obstetric patients; propofol (10 to 20 mg IV) had no effect on the onset or severity of pruritus after intrathecal morphine. Age and sex differences among these studies may account for the variable effectiveness of the similar doses of propofol used to mitigate neuraxial morphine-induced pruritus. These studies further underscore that pruritus disproportionately occurs in younger and female obstetric patients,^16,24–26  who may be less responsive to the antipruritic effects of propofol. Another possibility for the observed differences across studies is that the dose of propofol used may not have been adequate to achieve clinical effect. Thus, although propofol has been observed to reduces incidence of pruritus in patients receiving morphine for a variety of elective surgeries,^116.160 its benefit in the treatment of opioid-induced pruritus within the obstetric population remains questionable.¹¹⁷ 

Recent work in rodents have revealed that neurons producing GABA are required for neuraxial opioid–induced itch,^36,45  and the administration of propofol, through potentiation of GABA receptors, may enhance the ability of these neurons to dampen excitatory spinal circuits⁹²  involved in itch transmission. Additional dose-response studies are necessary to identify the appropriate dose of propofol for management of pruritus without sedation.

Gabapentin has been shown to be effective for several forms of pruritus in systemic disease, including uremic pruritus, pruritus in multiple sclerosis, and pruritus of unknown origin.^118,119,161  Its use in the treatment of neuraxial opioid–induced pruritus is less clear. Preoperative gabapentin (1,200 mg per os) has been found to significantly delay the onset and reduce the incidence and severity of intrathecal morphine–induced pruritus in patients undergoing orthopedic surgery¹⁶²  and prolong the onset of pruritus in patients receiving spinal morphine for unilateral hernia repair.¹²⁰  However, a lower dose (600 mg) was not found to significantly decrease the incidence of pruritus compared to placebo and was associated with urinary retention.¹⁶³  Further pharmacologic studies in preclinical and clinical models are necessary to characterize the mechanism of action of gabapentin within the nervous system. However, recent efforts have noted the safety concerns associated with gabapentin in the perioperative setting; the risks of adverse effects and lack of significant analgesic effect on postoperative pain suggest that the use of gabapentin for opioid-induced pruritus should also be avoided.^121,164 

Droperidol (1.25 mg IV), a short-acting, potent dopamine receptor antagonist, has been shown in several studies to be effective at reducing the incidence and severity of neuraxial opioid–induced pruritus and has previously been used for postoperative nausea and vomiting.^104,122,165  However, dopamine receptor antagonists have been reported to be effective only when small doses of opioids are administered. Sedation has been shown to increase with escalating doses of droperidol (2.5 to 5 mg IV), which may confound its antipruritic effects.¹²²  Given the broad influences of supraspinal dopamine signaling on the functions of spinal and dorsal root ganglion neurons, the antipruritic effects observed may be nonspecific.¹²³  Emerging evidence indicates that low doses of droperidol may be safely used; however, clinical use of droperidol in a perioperative setting is limited by an Food and Drug Administration black box warning because of its risk for sudden cardiac death.¹⁶⁶  The warning will likely preclude any clinical adoption of droperidol to treat opioid-induced pruritus.

All existing treatment options for opioid-induced pruritus have undesirable side-effect profiles. Furthermore, many of these treatments are used empirically and off-label. Because of the limitations of currently available treatments, some patients or providers may elect to forgo analgesia or opt for suboptimal analgesia that excludes neuraxial opioids, but these practices can lead to unnecessary pain and suffering.¹⁶⁷  Recent success in preclinical models suggests that nalfurafine, a selective κ-opioid receptor agonist, may be used to manage opioid-induced pruritus without limiting opioid-induced analgesia,³⁶  but further basic, translational, and clinical research is required before recommendations for clinical use in opioid-induced pruritus can be made. Optimization of opioid analgesia and development of improved therapies for opioid-induced pruritus have the potential to significantly improve the clinical standards of care for patients who receive opioids for the management of perioperative pain and experience pruritus as an unwanted side effect.

Pruritus after neuraxial opioids remains a highly common and dissatisfying side effect. Advancements in the understanding of mast cell biology and neuronal itch circuitry have provided clues as to how opioids can induce analgesia and evoke pruritus. Existing evidence suggests that parenteral opioids cause pruritus through histamine release, whereas neuraxial opioid–induced pruritus occurs through a mechanism of neuronal disinhibition in the spinal cord dorsal horn. The differential modulation of peripheral and neuraxial opioid–induced pruritus by mast cells and neurons, respectively, further highlight the complexity of the side effects of opioid use. Emerging evidence suggests that pruritus arises because of the dysregulation of opioid-sensitive pathways involving μ- and κ-opioid receptor signaling, which parallels other forms of chronic, systemic, and drug-induced pruritus. Ultimately, a richer understanding of the genetic, molecular, and cellular underpinnings of opioid-induced pruritus may provide a basis upon which to develop improved therapies that can manage pain but do not cause itch.

Supported by the Virginia Kaufman Endowment Fund; National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health (Bethesda, Maryland) grant No. AR063772; National Institute of Neurological Disorders and Stroke, National Institutes of Health grant No. NS096705 (to Dr. Ross); National Research Service Award grant No. F31 F31NS113371; and National Institute of General Medical Sciences, National Institutes of Health grant No. T32GM008208 (to Dr. Nguyen).

The authors declare no competing interests.