The Role of Peripheral N‐Methyl‐D‐Asparate (NMDA) Receptors in Itch and Pain: A Narrative Review

ABSTRACT

Background and Objective

NMDA receptors, predominantly located in the central nervous system and known for their roles in synaptic plasticity and central sensitisation of pain and itch, are also expressed in peripheral sensory neurons. Emerging evidence suggests that peripheral NMDA receptors contribute to pathological pain and potentially itch, identifying them as promising therapeutic targets. The aim of this review is to explore the role of peripheral NMDA receptors in pain and itch and to summarise the effectiveness of topical NMDA antagonists in managing these sensations.

Databases and Data Treatment

This review was conducted through a systematic search of the PubMed database using MeSH terms and keywords related to ‘peripheral NMDA receptors’, ‘pain’, ‘itch’ and ‘topical ketamine’. Additional references were included based on expert knowledge and reference tracking. Only English‐language articles were considered.

Results

Animal studies demonstrate that peripheral NMDA receptors are involved in inflammatory and certain neuropathic pain models, with antagonists showing analgesic effects. Limited studies also suggest their role in non‐histaminergic itch through glutamate signalling. In humans, topical ketamine has shown mixed results for pain relief, and preliminary clinical reports suggest potential antipruritic effects. However, controlled clinical trials, particularly for itch, are lacking.

Conclusion

Peripheral NMDA receptors are involved in the transmission and sensitisation of both pain and itch, especially under pathological conditions. While topical ketamine may offer therapeutic benefits, particularly for non‐histaminergic itch and neuropathic pain, further clinical research is necessary to confirm its efficacy, safety and optimal use.

Significance Statement

This review highlights the underexplored role of peripheral NMDA receptors in itch, especially non‐histaminergic pathways, and synthesises emerging evidence for their therapeutic potential. By comparing pain and itch mechanisms, it suggests that topical ketamine could serve as novel treatments for refractory itch and localised pain, warranting further clinical investigation.

1. Introduction

The NMDA receptor is a type of ionotropic glutamate receptor that functions as a ligand‐gated channel and mediates a major component of excitatory neurotransmission. The NMDA receptors are especially expressed in the central nervous system and play an important role in neuronal development, learning, memory and synaptic plasticity (Huang and Xiong 2021). In relation to peripheral signalling and modulation of pain and itch, NMDA receptors in the spinal cord are well known to contribute to central sensitisation. Continuous input from the peripheral afferent neurons depolarises the post‐synaptic cells of the spinal cord, resulting in activation of NMDA receptors. This leads to prolongation of increased excitability of the spinal neurons, which is seen in pathological pain and itch conditions (Basbaum et al. 2009; Cevikbas 2011).

Although the majority of NMDA receptors are located in the central nervous system, some are also expressed peripherally. Notably, presynaptic NMDA receptors were initially described by Liu et al. (1994), who provided anatomical evidence for autoreceptors in the spinal cord dorsal horn, and later demonstrated functional regulation of neurotransmitter release via these receptors (Liu et al. 1997). While these studies focused on central mechanisms, they laid the foundation for exploring NMDA receptor function beyond the synapse. Subsequent work identified NMDA receptors in peripheral sensory neurons, including their terminals, suggesting a potential role in transducing pain from various aetiologies (Carlton 2009; Petrenko et al. 2003). It is known that pain and itch share some common pathways. Pathological itch is classified into four types, which are pruriceptive, neuropathic, neurogenic and psychological (Liu and Ji 2013). This mimics the similar phenomena as seen for pain, which includes among others nociceptive, neuropathic and nociplastic pain. Up to now, there is a limited number of studies on peripheral NMDA receptors and itch, although it may be a future target for managing itch. The aim of the present review is to provide an overview of the similarities and differences in the peripheral neural NMDA‐related pathways of itch and pain. Key findings will be summarised with respect to the potential role of NMDA receptors in the peripheral signalling and modulation of pain and itch sensations, and the effectiveness of topical NMDA antagonists on pain and itch is summarised.

2. Methods

References for the present review were identified through a systemic search of the PubMed database regarding the relationship between peripheral NMDA receptors and the sensations of itch and pain. The search strategy included the use of Medical Subject Headings (MeSH) terms and relevant keywords to capture a broad range of studies on the role of peripheral NMDA receptors on itch and pain. The following search terms were used; ‘(“NMDA receptor”[All Fields] OR “N‐methyl‐D‐aspartate receptor”[All Fields]) AND (“pruritus”[MeSH Terms] OR “pruritus”[All Fields] OR “itch”[All Fields])’, ‘(“peripheral N‐methyl‐D‐aspartate receptor”[All Fields] OR “peripheral NMDA receptor”[All Fields] OR “peripheral N‐methyl‐D‐aspartate receptors”[All Fields] OR “peripheral NMDA receptors”[All Fields]) AND (“pain”[MeSH Terms] OR “pain”[All Fields])’, ‘(“topical”[All Fields] OR “topically”[All Fields] OR “topicals”[All Fields]) AND (“esketamine”[Supplementary Concept] OR “esketamine”[All Fields] OR “ketamine”[All Fields] OR “ketamine”[MeSH Terms] OR “ketamin”[All Fields] OR “ketamine s”[All Fields] OR “ketamines”[All Fields]) AND (“pruritus”[MeSH Terms] OR “pruritus”[All Fields] OR “itch”[All Fields])’.

We also employed reference tracking. All identified studies were screened for inclusion. In addition to database searching, articles which are found based on our existing knowledge and resources are included (Greenhalgh and Peacock 2005). Only articles in English have been included.

2.1. Similarities and Distinctions in Peripheral Neural Pathway of Itch and Pain

Both itch and pain are unpleasant and warning sensations. Itch evokes a desire to scratch to remove irritants, whereas acute pain provokes a withdrawal response. These facts suggest that itch and pain are distinct sensations with distinct different roles for survival. However, itch and pain have similarities in sensory transmission at the primary afferent neuronal level. Both itch and pain utilise unmyelinated C fibres and myelinated Aδ fibres for transmitting their sensations (Yosipovitch et al. 2020). Chemical itch is mediated by C fibres, whereas mechanical itch is conveyed by Aδ fibres. Additionally, secondary hyperalgesia and hyperknesis are attributed to central sensitisation to input from Aβ afferents (Andersen et al. 2018; Yosipovitch et al. 2020). There are a number of overlapping mediators involved in the transmission of itch and pain sensations, such as amines (e.g., serotonin, histamine) (Dong and Dong 2018; Obara et al. 2020), peptides (e.g., bradykinin, endothelin‐1) (Hosogi et al. 2006; McNeil and Dong 2012), sphingolipids (e.g., sphingosine 1‐phosphate) (Hill et al. 2018) and cytokines (e.g., IL‐33, thymic stromal lymphopoietin) (Gao et al. 2023; Yosipovitch et al. 2024). These mediators can exert their effects either directly on free nerve endings and keratinocytes or indirectly by enhancing the activity of other mediators (Yosipovitch et al. 2024). Pain and itch also share ion channels for signal transmission, including the voltage‐gated sodium channels (e.g., Nav1.7, Nav1.8 and Nav1.9) and transient receptor potential (TRP) cation channel subfamily (e.g., TRPV1 and TRPA1) (Liu and Ji 2013; Yosipovitch et al. 2024). In addition, some receptors that induce itch, such as protease‐activated receptors (PARs) and Mas‐related G protein‐coupled receptors (Mrgpr), are shown to be involved in forming pathological conditions of pain (Vergnolle et al. 2001; Green 2021).

How the somatosensory system distinguishes itch from pain to initiate the correct response, such as removal (fight) or withdrawal (flight), has been a subject of debate for over a century, with the understanding of these neural mechanisms continuing to develop. There have been several theories proposed, such as specificity theory (labelled line theory), intensity theory, selectivity theory, leaky gate theory and spatial contrast theory (Okutani et al. 2024; Chen 2021). Among these theories, specificity theory is the most studied. This theory posits that primary afferent fibres, which are specialised for signalling either itch or pain, connect directly to specific central pathways for each sensation. However, neurons that are activated by pruritogens also respond to nociceptive agents such as capsaicin, mustard oil and other noxious stimuli. There is no evidence for peripheral sensory neurons that are specifically tuned to itch alone, responding only to pruritic stimuli and not to nociceptive ones (LaMotte et al. 2014). Contrary to the labelled line theory, it has been shown that primary afferents can initiate somatosensory discrimination through cell‐autonomous processes at the peripheral level (Sharif et al. 2020). For instance, metabotropic activation of MrgprA3 C‐afferents, which has been suggested as a specific pathway for itch (Han et al. 2013), primarily induces itching, whereas rapid ionotropic activation of the same group of neurons mainly causes pain (Sharif et al. 2020). Therefore, it is valid to think that agents that act on peripheral neurons to alleviate pain are also effective in the attenuation of itch.

2.2. NMDA Receptor: Definition, Subunit Composition and Functional Properties

The NMDA receptor is a type of ionotropic glutamate receptor that functions as a ligand‐gated channel, allowing the passage of cations such as sodium (Na⁺), potassium (K⁺) and calcium (Ca²⁺) (Kreutzwiser and Tawfic 2019). NMDA receptors exist as heterotetramers consisting of two obligately GluN1 subunits in combination with two GluN2A‐B and/or GluN3A‐B subunits (Petrenko et al. 2014; Vyklicky et al. 2014). GluN2 and GluN3 subunits contribute significantly to the functional diversity of NMDA receptors through their varying incorporation (Kreutzwiser and Tawfic 2019; Cull‐Candy et al. 2001). For example, a brief application of glutamate to GluN1/GluN2A assemblies induces a macroscopic current with a deactivation time constant on the order of tens of milliseconds. In comparison, GluN1/GluN2D receptors show a deactivation time constant that extends over several seconds (Cull‐Candy et al. 2001).

NMDA receptors are found predominantly in the central nervous system and play important roles in synaptic development, synaptic plasticity, hyperexcitability, neuroplasticity, learning, memory and cognition (Huang and Xiong 2021). NMDA receptors have specific features distinct from other ionotropic glutamate receptors such as kainate receptors and α‐amino‐3‐hydroxy‐5‐methyl‐4‐isoxazolepropionic acid (AMPA). For efficient activation of NMDA receptors (NMDAR), both glutamate and glycine must bind simultaneously. At the resting membrane potential, NMDAR channels are obstructed by extracellular magnesium. This magnesium block is removed when the membrane potential undergoes sufficient depolarisation in both amplitude and duration (Kreutzwiser and Tawfic 2019). NMDAR produces a current characterised by a gradual rise and an exceptionally slow decay. The channels typically open around 10 ms after glutamate is released into the synaptic cleft, and they remain open and close intermittently for hundreds of milliseconds until glutamate dissociates from the receptor. NMDA receptors are highly permeable to Ca²⁺, and the resulting influx of intracellular Ca²⁺ initiates a cascade of downstream signalling events (Kreutzwiser and Tawfic 2019).

NMDA receptors have also been identified in the cell bodies, as well as in the peripheral and central processes of primary sensory neurons in both rodents and humans (Carlton 2009). Glutamate is likely released from the peripheral terminals of primary afferent C fibres, as it has been demonstrated immunocytochemically in a subpopulation of dorsal root ganglion cells, and these neurons are known to release glutamate from their central terminals (Salt and Hill 1983). This suggests that peripheral glutamate release may mirror mechanisms occurring centrally, enabling the activation of NMDA receptors in the periphery (Omote et al. 1998). Additionally, non‐neuronal sources such as macrophages may also contribute to the local glutamate pool, particularly under inflammatory conditions (Piani et al. 1991). Further supporting the role of glutamate signalling in non‐neuronal tissues, Hinoi et al. demonstrated that in various peripheral tissues, glutamate can function as an autocrine or paracrine signalling molecule. In these contexts, cells both release and respond to extracellular glutamate, thereby modulating their own activity or that of neighbouring cells. Notably, NMDA receptors expressed in peripheral tissues play a critical role in mediating these glutamate‐induced responses (Hinoi et al. 2004). In the periphery, approximately 50% of the myelinated fibres including Aδ and Aβ and 20%–30% of the unmyelinated C fibres are labelled for GluN1 (Carlton 2009). Peripheral NMDA receptors are present in the dermal–epidermal junction (Carlton et al. 1998). Peripheral nociceptive fibres are known to express GluN2B and GluN2D subunits, while GluN2A subunits are absent from the peripheral terminals of primary afferents (Petrenko et al. 2003). The GluN2B subunit is co‐localised with markers of small diameter and/or nociceptive DRG cells, such as wheat germ agglutinin horse radish peroxidase (WGA‐HRP), isolectin B4 (IB4) and calcitonin gene‐related peptide (CGRP) (Ma and Hargreaves 2000). Half of the NMDA‐expressing fibres in the temporalis nerve also co‐expressed substance P and CGRP (Gazerani, Au, et al. 2010; Gazerani, Dong, et al. 2010). Many NMDA receptors act as autoreceptors, influencing glutamate release from primary afferents and modulating the release of CGRP and substance P (Liu et al. 1994; McRoberts et al. 2001), as demonstrated by Liu et al. (1997), who showed that activation of presynaptic NMDA receptors enhances neurotransmitter release from primary afferent terminals. Substance P amplifies the activity induced by glutamate and NMDA in the spinal cord, with this amplification playing an important role in the development of central sensitisation. It is proposed that primary afferent neurons may undergo similar mechanisms as those in the spinal cord, leading to prolonged depolarising potentials (Carlton et al. 1998). It should be noted that the peripheral NMDA facilitation and presynaptic mechanisms remain a matter of ongoing debate. Further research is warranted to clarify the physiological relevance and mechanistic underpinnings of these peripheral presynaptic NMDA receptor functions.

2.3. The Potential Role of Peripheral NMDA Receptors in Pain and Itch: Insights From Animal and Human Studies

2.3.1. Animal Studies

Glutamate acts as a nociceptive agent through the activation of NMDA receptors. Glutamate administered externally activates peripheral glutamate receptors, leading to the depolarisation of primary afferent C fibres (Evans et al. 1987) and the manifestation of pain‐related behaviours (Jackson et al. 1995; Leem et al. 2001; Du et al. 2003; Gazerani, Au, et al. 2010; Gazerani, Dong, et al. 2010). Glutamate‐induced mechanical sensitisation on skin afferent fibres is alleviated by a selective NMDA receptor antagonist (Gazerani, Au, et al. 2010; Gazerani, Dong, et al. 2010). Pain‐related behaviours are also evoked by intra‐articular and masseter injection of glutamate and attenuated by selective NMDA receptor antagonists (Lawand et al. 1997; Cairns et al. 2003). NMDA antagonists administered peripherally also attenuated monosodium glutamate‐induced headache (O'Brien and Cairns 2016; Benbow et al. 2022) and pain derived from colonic distention (McRoberts et al. 2001) and tooth movement (Yang et al. 2009). These results suggest that peripheral NMDA receptors also contribute to various kinds of somatic and visceral pain.

Table 1 summarises the effect of NMDA receptor antagonists administered peripherally on experimental animal models of pain. The knowledge of the role of peripheral NMDA receptors in pain from animal studies is primarily based on inflammatory pain, with only a few studies addressing neuropathic pain.

TABLE 1.

The effect of peripheral N‐methyl‐D‐asparate (NMDA) receptor antagonism on experimental pain models in animal studies.

Author and year	Noxious stimulus	Species	Antagonist/Intervention	Effect	Route of administration	Timing
Gazerani, Au, et al. (2010), Gazerani, Dong, et al. (2010)	Glutamate into the non‐glabrous skin of the face	Rat	AP5	Attenuation of a glutamate‐induced mechanical sensitisation of skin afferent fibres.	s.c.	Simultaneously
Lawand et al. (1997)	Glutammate into the knee joint cavity	Rat	AP7	Attenuation of thermal hyperalgesia and the mechanical allodynia.	Intra‐articular injection	Simultaneously
Cairns et al. (2003)	Glutamate or hypertonic into the masseter muscle	Rat	AP5, ketamine	Reduction of glutamate‐evoked afferent discharge in rats.	Injection in the masseter muscle
Benbow et al. (2022)	MSG‐induced headache	Rat	AP5	Attenuation of nocifensive and spontaneous headache‐like behaviour.	i.p.	Simultaneously
O'Brien and Cairns (2016)	MSG‐induced headache	Rat	AP5	Attenuation of increase in neuronal discharge and mechanical sensitisation of SpVc neurons.	Systemic injection	Simultaneously
McRoberts et al. (2001)	Colonic distention	Rat	Memantine	Inhibition of electromyographic activity in the eternal oblique musculature quantified as visceromotor response. Inhibition of the afferent nerve activity induced by colorectal distention.	i.v.	Prestimulation
Yang et al. (2009)	Experimental tooth movement	Rat	MK801	Reduction of nocifensive behaviour.	Injected into the periodontal ligament of the two lower incisors and first molars	Post‐treatment
Davidson et al. (1997)	Formalin into hindpaw	Rat	MK801	Attenuation of lifting/licking behaviour during phase 2.	s.c. in the hind toe	Pre‐treatment
Davidson and Carlton (1998)	Formalin into hindpaw	Rat	Dextorphan Ketamine Memantine	Dextorphan, ketamine and memantine decreased formalin‐induced lifting/licking behaviour in phase 2. Memantine also decreased that in phase 1.	i.pl.	Pre‐treatment
Christoph et al. (2005)	Formalin into hindpaw	Rat	5,7‐DCK	Attenuation of formalin‐evoked behaviour during phase 2.	i.v.	Pre‐treatment
Tan et al. (2011)	Formalin into hindpaw	Rat	A knockdown of the GluN1 subunit of peripheral NMDA receptor	Attenuation of a formalin‐induced nociceptive behaviour during phase 2.
Leem et al. (2001)	CFA into hindpaw	Rat	MK801	Attenuation of the mechanical hyperalgesia.	i.pl.	Post‐treatment
Du et al. (2003)	CFA into hindpaw	Rat	MK801	Attenuation of the mechanical hyperalgesia.	i.pl.	Post‐treatment
Jackson et al. (1995)	Carrageenan into hindpaw	Rat	MK801	Attenuation of the thermal hyperalgesia.	i.pl.	Post‐treatment
Chen et al. (1999)	Bee venom‐induced inflammation of hindpaw	Cat	AP5	Both pre‐ and post‐treatment with AP5 suppressed the increased firing of the WDR neurons of the spinal cord.	s.c. in the hind paw	Pre‐ and post‐treatment
You et al. (2002)	Bee venom‐induced inflammation of hindpaw	Rat	AP5 MK801	Pre‐treatment with AP5 and both pre‐treatment and post‐treatment with MK801 reduced the bee venom‐increased spontaneous responses of WDR neuron of the spinal cord. AP5 reduced mechanical allodynia and hyperalgesia.	i.pl.	Pre‐ and post‐treatment
Romero et al. (2011), Romero and Duarte (2013)	Prostaglandin E2 into hindpaw	rat	Ketamine	Attenuation of the mechanical hyperalgesia.	i.pl.	Post‐treatment
Oatway et al. (2003)	Mild thermal injury on the hindpaw	Rat	Ketamine	Attenuation of the thermal hyperalgesia.	i.pl.	Pre‐treatment
Lawand et al. (1997)	Injection of kaolin and carrageenan into the knee joint cavity	Rat	AP7 Ketamine	Attenuation of thermal hyperalgesia and the mechanical allodynia.	Intra‐articular injection	Simultaneously
Yu et al. (1996)	Mustard oil injection into the TMJ	Rat	MK801	Attenuation of the magnitudes of both the masseter and digastric EMG response. No significant effect on the plasma extravasation.	Injection into the TMJ	Pre‐treatment
Ro (2003)	Masseteric injection of mustard oil	Rat	MK801	Attenuation of the nocifensive behaviour as well as the formation of oedema in the masseter muscle.	Injection in the masseter muscle	Pre‐treatment
Ro et al. (2007)	Hypertonic saline infused in the masseter muscle	Rat	MK801	Attenuation of the hypertonic saline induced nocifensive behaviour. Reduction of c Fos positive neurons induced by hypertonic saline in the ipsilateral subnucleus caudalis	Injection in the masseter muscle	Pre‐treatment
Ivanusic et al. (2011)	CFA injection into the TMJ	Rat	AP5 Ifenprodil	Attenuation of the mechanical hypersensitivity.	Injection into the TMJ	Simultaneously
Wong et al. (2014)	NGF into the masseter muscle	Rat	AP5	Reversal of the mechanical sensitisation in male but not in female.	Injection into the masseter muscle	Post‐treatment
Jin et al. (2009)	Capsaicin into the hindpaw	Rat	MK801	Attenuation of the thermal hyperalgesia. Attenuation of capsaicin‐induced glutamate release.	s.c. into the hindpaw	Simultaneously
Lam et al. (2005)	Capsaicin injection into TMJ	Rat	MK801 AP5	Reduction in the magnitude of capsaicin‐evoked digastric and masseter EMG activity in a concentration‐related manner.	Injection into the TMJ	Pre‐treatment
Aley and Levine (2002)	Neuropathic pain (Streptozotocin, Vincristine, CCI, PSL)	Rat	MK801	MK801 did not have a significant effect on mechanical hyperalgesia induced by all four models of neuropathic pain.	Intradermally into the dorsum of the hindpaw	Post‐treatment
Jang et al. (2004)	L5 SNL following L5 dorsal rhizotomy	Rat	MK801	Reduction in mechanical hyperalgesia.	i.pl.	Pre‐ and post‐treament
Christoph et al. (2005)	Neuropathic pain (CCI, SNL)	Rat	5,7‐DCK	Attenuation of cold allodynia in the CCI model and tactile allodynia in the SNL model.	i.v.	Post‐treatment
Xu et al. (2020)	CPIP	Rat	Ifenprodil	Attenuation of mechanical allodynia and thermal hyperalgesia both in acute and chronic phase of CPIP.	s.c. in the hindpaw	Pre‐treatment (acute phase) Post‐treatment (choronic phase)

Note: Unless stated otherwise, the antagonist was administered at the same site as the pain‐inducing area.

Abbreviations: CCI, chronic constriction injury; CFA, Complete Freund's adjuvant; CPIP, chronic post‐ischaemic pain; i.p., intraperitoneally injection; i.pl., intraplantar injection; i.v., intravenous injection; MSG, monosodium glutamate; PSL, partial sciatic nerve ligation; s.c., subcutaneous injection; SNL, spinal nerve ligation; SpVc, spinal trigeminal subnucleus caudalis; TMJ, temporomandibular joint; WDR, wide dynamic range.

NMDA receptor antagonists are effective in providing pain relief for inflammatory pain induced by various inflammatory mediators and thermal injury. Peripheral NMDA receptors are involved in the process of peripheral sensitisation during inflammatory responses. The increase in glutamate release in the periphery is observed by various stimulations such as formalin injection (Omote et al. 1998) and melitine (Iwashita et al. 2012). Peripheral NMDA receptor is associated with neurogenic inflammation, shown by the fact that peripherally administered NMDA receptor antagonist attenuated flare and formation of oedema (Ro 2003; Iwashita et al. 2012), though a controversial result is also suggested (Yu et al. 1996). Importantly, foundational studies in the central nervous system by Liu et al. (1997) demonstrated that presynaptic NMDA receptor activation enhances the release of substance P, which is a key mediator of neurogenic inflammation. These findings support the plausibility that similar presynaptic mechanisms operate in the periphery, contributing to inflammation through the modulation of neuropeptide release.

In inflammatory conditions, the proportions of myelinated and unmyelinated axons labelled for NMDA receptors increased to 61% in cutaneous nerves in the inflamed paw, compared to about 40%–50% in the non‐inflamed paw (Carlton and Coggeshall 1999). Peripheral administration of NMDA receptor antagonists was shown to be effective in various models of inflammatory pain. The antagonist MK801 attenuates the formalin‐induced pain behaviour during phase 2, which is caused by inflammatory aetiology, although it has no effect on that during phase 1, which is the acute nociceptive response (Davidson et al. 1997). Although the antagonist was ineffective during the initial acute phase, glutamate release has been shown to occur immediately following formalin injection (Omote et al. 1998). This suggests that NMDA receptors in the periphery are activated early, but may not play a central role in mediating the initial nociceptive response. Instead, they may contribute more significantly to the amplification and maintenance of pain during the inflammatory phase, as evidenced by the increased expression of NMDA receptors under inflammatory conditions (Carlton and Coggeshall 1999) and the selective attenuation of phase 2 pain behaviours by NMDA antagonists (Davidson et al. 1997). MK801 is effective on thermal and mechanical hyperalgesia induced by carrageenan (Jackson et al. 1995) and Freund's complete adjuvant (Leem et al. 2001; Du et al. 2003). Peripheral NMDA receptors also contribute to inflammatory pain in muscle and knee joint, indicated by the fact that selective NMDA antagonists injected in muscle and knee joint attenuates these pain (Lawand et al. 1997; Ivanusic et al. 2011).

Administering NMDA antagonists to the periphery might offer a viable treatment for neuropathic pain, although its efficacy in neuropathic pain is controversial. Christoph et al. (2005) showed that the NMDA antagonist, 5,7‐DCK, which has limited access to the central nervous system, reversed allodynia due to two neuropathic pain models, the chronic constriction injury of the sciatic nerve and the L5/6 spinal nerve ligation. In addition, MK‐801 was shown to attenuate mechanical hyperalgesia in a rat model of neuropathic pain induced by L5 spinal nerve lesion (SNL) proceeded by L5 dorsal rhizotomy, which is devoid of the potential central effects induced by L5 SNL‐induced impulses through the L5 dorsal root (Aley and Levine 2002). Peripheral administration of NMDA antagonists has also been shown to relieve chronic post‐ischaemic pain (Xu et al. 2020). However, MK‐801 is not effective on traumatic peripheral neuropathy pain induced by two models of neuropathic pain, chronic constriction and partial injury of the sciatic nerve. MK‐801 is also shown to be ineffective on models of diabetic/chemotherapeutic neuropathic pain (Jang et al. 2004).

Animal research of NMDA receptors in itch is very limited. It is reported that the activation of NMDA receptors has a relation to non‐histaminergic itch. Haddadi et al. (2017) reported that the NMDA/NO pathway in the skin modulates chloroquine (CQ)‐induced itch, which is a non‐histaminergic itch evoked via activation of a G protein‐coupled receptor termed Mrgprs (MrgA3) on C‐fibres. They showed that the peripheral NMDAR antagonist, MK‐801, significantly decreased CQ‐induced itch along with the intradermal nitrite level, which is the end product of NO metabolism and increases by intradermal administration of CQ. The mechanism of action is that the TRPA1 channel, which is coupled to MrgprA3 via G_βγ may activate the glutamatergic system and NMDA receptors, leading to the production of NO, which acts as a neurotransmitter in itch (Haddadi et al. 2017).

2.3.2. Human Studies

Glutamate also acts as a nociceptive agent in humans (Cairns et al. 2003, 2006; Castrillon et al. 2007, 2012; Alstergren et al. 2010). The relationship between peripheral NMDA receptors and pain in humans has been studied using the capsaicin model and the thermal injury model (Table 2). Capsaicin evokes pain via activation of TRPV1. The TRPV1 receptor is shown to be co‐localised with peripheral NMDA receptors on the same primary afferent terminal, and capsaicin induces glutamate release at the peripheral terminals of primary afferents (Jin et al. 2009). The NMDA antagonist MK801 attenuates peripheral glutamate release and thermal hyperalgesia in a pain model of a rat injected with capsaicin in its hind paw (Jin et al. 2009). MK801 also attenuates digastric and masseter EMG activity evoked by injection of capsaicin into the temporomandibular joint (TMJ) of rats (Lam et al. 2005). As described in the previous section, NMDA antagonists have an analgesic effect on animals; however, in humans it has been shown that there is no evidence for the involvement of peripheral NMDA receptors in capsaicin‐induced pain and hyperalgesia in studies using ketamine as an NMDA antagonist (Koppert et al. 1999; Gottrup et al. 2000, 2004; Pöyhiä and Vainio 2006). These results contrast with those showing that ketamine attenuated pain caused by a mild heat injury (Warncke et al. 1997; Pedersen et al. 1998). These differences are discussed in terms of the involvement of different types of neurons (Warncke et al. 1997): the former being mechano‐insensitive C fibres (CMi) (LaMotte et al. 1982), and the latter being mechano‐heat‐sensitive C fibres (CMH) (Yosipovitch et al. 2020).

TABLE 2.

The effect of peripheral N‐methyl‐D‐aspartate (NMDA) receptor antagonism on experimental pain models in human studies.

Author and year	Noxious stimulus	Antagonist/Intervention	Effect	Route of administration	Timing
Cairns et al. (2003)	Glutamate or hypertonic into the masseter muscle	Ketamine	Reduction of glutamate‐evoked muscle pain in human.	Injection in the masseter muscle
Cairns et al. (2006)	Glutamate into the masseter muscle	Ketamine	Attenuation of the glutamate‐induced mechanical sensitisation.	Injection in the masseter muscle	Simultaneously
Castrillon et al. (2007)	Glutamate into the masseter muscle	Ketamine	No effect on the glutamate‐induced mechanical sensitisation.	Injection in the masseter muscle	Simultaneously
Castrillon et al. (2012)	Glutamate into the masseter and temporalis muscle	Ketamine	Attenuation of glutamate‐induced pain.	Injection in the masseter and temporalis muscle	Simultaneously
Alstergren et al. (2010)	Glutamate into the TMJ	Ketamine	Attenuation of glutamate‐induced pain	Injection into the TMJ	Simultaneously
Warncke et al. (1997)	Thermal injury (1° burn injury) on the medial surface of a calf	Ketamine (4.98 mg/6 mL)	Inhibition of primary and secondary hyperalgesia.	s.c. in the same skin area as burn‐injured area	Pre‐treatment
Pedersen et al. (1998)	Thermal injury (47° burn injury) on the medial part of the distal leg	Ketamine (7.5 mg/5 mL)	Reduction in pain intensity. Attenuation of the thermal and mechanical hyperalgesia. Secondary hyperalgesia and suprathreshold pain response to heat and mechanical pain stimuli were not affected.	Local infiltration in the burn area	Pre‐treatment
Koppert et al. (1999)	Intradermal injection of 10 μg/20 μL of capsaicin in the forearm	200 μL ketamine solution (100 or 1000 μg, respectively)	1000 μg of ketamine, which is a local anaesthetic dose attenuated the intensity of secondary hyperalgesia and axon reflex flare, while ketamine in the smaller concentration did not.	Intradermal injection in the forearm	Pre‐treatment
Gottrup et al. (2000)	Intradermal injection of 100 μg/20 μL of capsaicin in the forearm	Ketamine (5.0 mg/2.0 mL)	No effect on spontaneous pain, mechanical evoked pain, area of hyperalgesia.	Intradermal injection in the forearm	Pre‐treatment
Gottrup et al. (2004)	Intradermal injection of 100 μg/20 μL of capsaicin in the forearm	Ketamine (5.0 mg/2.0 mL)	Reduction of the flux areas both in the primary and hyperalgesic area. No effect on spontaneous pain, mechanical evoked pain, area of hyperalgesia.	Intradermal injection in the forearm	Pre‐treatment
Pöyhiä and Vainio (2006)	Intradermal injection of 250 μg/25 μL of capsaicin in the forearm	Ketamine gel (50 mg/1 mL)	Attenuation of the mechanical hyperalgesia. However, this effect was also observed when ketamine was administered in the contralateral side to the capsaicin injected side. Therefore, authors concluded that this effect would be due to the systemic effect of ketamine.	Rubbed into the skin of the forearm	Pre‐treatment

Note: Unless stated otherwise, the antagonist was administered at the same site as the pain‐inducing area.

Abbreviations: s.c., subcutaneous injection; TMJ, temporomandibular joint.

In contrast to research on NMDA receptors in human pain, no human studies have been conducted so far on the effect of peripheral NMDA receptor blocking on itch transduction and transmission. The itch pathway consists of two mechanisms: the histamine‐dependent pathway and the histamine‐independent pathway, classified based on whether histamine is the mediator or not (Yosipovitch et al. 2020). The histamine‐dependent pathway is transmitted by CMi fibres, whereas the histamine‐independent pathway is transmitted by CMH fibres (Yosipovitch et al. 2020). Based on findings from studies using human pain models, the role of peripheral NMDA receptors in histaminergic itch may be smaller than in non‐histaminergic itch. Further studies are needed to confirm the role of peripheral NMDA receptors in itch.

In summary, based on experiments using inflammatory pain models in animals, peripheral NMDA receptors on both myelinated and unmyelinated neurons are upregulated under inflammatory conditions, leading to the enhancement of pain behaviours. The fact that capsaicin‐evoked pain is not alleviated by peripheral NMDA receptor blockade in humans indicates that these receptors have little involvement in CMi fibres, which are also responsible for conveying histaminergic itch. Accordingly, we speculate that their contribution to histaminergic itch in humans may also be minimal. In contrast, peripheral NMDA receptor blockade is effective against pain caused by mild heat injury, which is transmitted by CMH fibres. CMH fibres also play a critical role in conveying non‐histaminergic itch. Considering the results demonstrating the efficacy of peripheral NMDA antagonists in alleviating non‐histaminergic itch, it is likely that peripheral NMDA receptors on CMH fibres play an important role in the sensitisation of both nociceptors and pruriceptors (Figure 1).

FIGURE 1.

Speculated role of NMDA receptors on C fibres for transduction of pain and itch. Mechano‐insensitive C (CMi) afferents are responsible for the transduction of histaminergic itch and capsaicin‐evoked pain. The fact that capsaicin‐evoked pain is not alleviated by peripheral NMDA receptor blockade in humans indicates that these receptors have little involvement in CMi fibres. Accordingly, we speculate that their contribution to histaminergic itch in humans may also be minimal. Therefore, we omitted NMDA receptors on CMi fibres. On the other hand, peripheral NMDA receptor blockade is effective against chloroquine‐induced, non‐histaminergic itch and pain caused by mild heat injury, which is transmitted by mechano‐heat‐sensitive C (CMH) fibres. It is likely that peripheral NMDA receptors on CMH fibres play an important role in the sensitisation of both nociceptors and pruriceptors.

2.4. Clinical Efficacy of Peripheral NMDA Receptors Antagonist on Pain and Itch: Are Peripheral NMDA Receptors an Attractive Target for Itch and Pain Treatment?

Up to now, selective NMDA antagonists for topical application are not commercialised. Ketamine, which predominantly acts as an NMDA antagonist, can be administered topically in the form of cream or gel. In the clinical setting, topical ketamine is mainly used as an analgesic and has been reported to relieve various pathogenic forms of pain, especially in localised neuropathic pain (Sawynok 2014). Ketamine cream is used alone or in combination with other analgesics such as amitriptyline, baclofen, lidocaine, ketoprofen, clonidine and gabapentin. As reviewed by Sawynok (2014), clinical trials and case reports showed that topical ketamine in combination with other agents, particularly amitriptyline, alleviates neuropathic pain. Topical ketamine alone has been reported to alleviate neuropathic pain (Sawynok 2014) and ketamine oral rinse provides effective palliation of intractable mucositis pain of the tongue (Slatkin and Rhiner 2003) in case reports. Therefore, peripheral action of ketamine alone could be a potential treatment for pain, though these findings have not been consistently replicated in controlled trials (Sawynok 2014). Topical racemic ketamine cream or gel, in concentrations up to 20%, has been effective for managing chronic pain with few reported side effects (Kopsky et al. 2015). The optimal dosages and drug combinations, however, are still yet to be determined.

The evidence supporting antipruritic actions of topical ketamine consists of case reports and retrospective studies without control (Table 3). Previous retrospective studies, which included patients with various pathological itches, showed that 55%–63% of patients had relief of itch after applying topical ketamine (Lee et al. 2017; Poterucha, Murphy, Davis, et al. 2013; Poterucha, Murphy, Sandroni, et al. 2013). The extent of the effect of topical ketamine seems to be above 50% reduction of itch numerical rating scale (Magazin et al. 2019; Jaller and Yosipovitch 2018; Poterucha, Murphy, Davis, et al. 2013; Poterucha, Murphy, Sandroni, et al. 2013). However, topical ketamine was used in combination with other agents such as amitriptyline and lidocaine. It is uncertain if the effect of topical ketamine on itch is attributed to ketamine alone. Side effects were reported in 0%–16.7% of patients, and most of these are local reactions such as burning sensation and redness (Lee et al. 2017; Mirzoyev and Davis 2013). Therefore, it seems to have a good tolerability to use. However, toxic encephalopathy is reported in a patient with Parkinson disease and atopic dermatitis after application of ketamine 10%, amitriptyline 5% and lidocaine 5% due to systemic absorption of ketamine, which is attributed to application in the setting of an impaired skin barrier (Cardis and Pasieka 2016). Patients who benefit from this treatment should be determined in further studies.

TABLE 3.

The effect of topical ketamine in combination with other agents on itch.

Author and year	Study design	n	Type of itch	Agent, dose	Results and comments
Ferreira et al. (2020)	A case report	1	Epidermolysis bullosa	Ketamine 0.5% and amitriptyline 2%	A significant reduction in itch within 6 weeks.
Magazin et al. (2019)	A case report	1	Brachioradial pruritus	Ketamine 0.5% and amitriptyline 1% 3–4 times daily	Itch NRS decreased from 6/10 to 2/10 within a day. No symptoms of pruritus 1 week later. Resolution of excoriations 5 weeks later.
Jaller and Yosipovitch (2018)	A case report	1	Inflammatory linear verrucous epidermal nevus (ILVEN)	Ketamine 10%, amitriptyline 5% and lidocaine 5% 1–3 times daily	Itch NRS decreased from 9.5/10 to 0/10 after 6 weeks.
Griffin and Davis (2015)	A case report	1	Post‐herpetic neuralgia	Ketamine 0.5% and amitriptyline 2%	Modest control of the pruritus and pain was achieved with continued multimodal therapy.
Poterucha, Murphy, Davis, et al. (2013), Poterucha, Murphy, Sandroni, et al. (2013)	A case report	1	Brachioradial pruritus	Ketamine and amitriptyline	The patient had complete symptom relief.
Lee et al. (2017)	A retrospective study without control	96	Brachioraidal pruritus (n = 9) Notalgia parethetica (n = 3) Neuropathic pruritus NOS (n = 16) Prurigo nodularis (n = 18) Atopic dermatitis (n = 12) Chronic pruritus NOS (n = 14) Other (n = 36)	Ketamine 10%, amitriptyline 5% and lidocaine 5% (TKAL) (n = 80) or ketamine 5%, amitriptyline 5% and lidocaine 5% (TKAL) (n = 16) up to 3 times daily.	63% of patients attributed relief of itch directly to the use of TKAL alone. The itch NRS decreased from 8 (1.62) 4.19 (2.9) after treatment with a reduction of 4.61 (2.77) (mean [standard deviation]).
Poterucha, Murphy, Davis, et al. (2013); Poterucha, Murphy, Sandroni, et al. (2013)	A retrospective study without control	16	Pruritus NOS (n = 5) Neurodermatitis (n = 4) Pruritus caused by post‐herpetic neuralgia (n = 2) Nostalgia paresthetica (n = 2) Anaesthesia dolorosa (n = 1) Nasal pruritus (n = 1) Diabetic neuropathy (n = 1)	Ketamine 0.5% and amitriptyline 2% or 1% 1–5 times daily for a mean duration of 10 months (range 1–52).	10 (63%) patients had relief of itch to some extent and 2 of them had complete resolution.
Mirzoyev and Davis (2013)	A retrospective study without control	11	Brachioradial pruritus	Ketamine and amitriptyline (dose were not mentioned)	6 (55%) patients had relief to some extent and 3 of them complete resolution.

Abbreviation: NRS, numerical rating scale.

3. Conclusion

Peripheral NMDA receptors are present on both nociceptive and purinergic (histaminergic and non‐histaminergic) afferents and may play a role in the development of various types of pathological pain and itch.

Blocking the peripheral NMDA receptor may be effective in reducing pain and itch transduction and transmission. Clinically, topical ketamine has emerged as a potential treatment for pain, especially neuropathic pain, although its optimal dosage and combination with other medications need to be determined through controlled trials. So far, no human studies have been done on the effect of peripheral NMDA receptor blocking on itch transduction and transmission.

So compared to research on pain involving peripheral NMDA receptors, studies on itch are still relatively scarce. Experimental research, however, indicates that peripheral NMDA receptors are involved in chloroquine‐induced, non‐histaminergic itch. The clinical effectiveness of blocking peripheral NMDA receptors for itch has been supported primarily by case reports, which have shown that topical ketamine could be effective for various pathological itch conditions with cutaneous manifestations, suggesting that peripheral NMDA receptors may contribute to itch pathology and hence a blockade could serve as an antipruritic treatment. However, further investigation is needed to clarify the precise role of peripheral NMDA receptors on itch in both experimental and clinical studies.

Tanaka, R. , Lo Vecchio S., Aliotta G. E., and Arendt‐Nielsen L.. 2025. “The Role of Peripheral N‐Methyl‐D‐Asparate (NMDA) Receptors in Itch and Pain: A Narrative Review.” European Journal of Pain 29, no. 7: e70082. 10.1002/ejp.70082.