Chronic Pain in Dogs and Cats: Is There Place for Dietary Intervention with Micro-Palmitoylethanolamide?

Pain includes at least two dimensions, i.e., physical and emotional components. From a physical perspective, although pain is often conceived as a homogeneous sensory entity, several distinct types exist: transient, inflammatory, neuropathic and functional pain ( Figure 1 ) [ 54 ]. Transient pain develops when a potentially harmful insult is applied to a superficial or deep tissue (cutaneous/mucous or musculoskeletal/visceral, respectively) for such a short time that it does not cause tissue damage (potential damage). It develops rapidly and has a transient nature, disappearing with the end of the harmful stimulus or shortly thereafter. Transient pain acts like an alarm signal, capable of activating a sudden withdrawal reflex that protects the tissues from the noxious stimulus (adaptive pain). It develops thanks to the activation of the nociceptive system and the transduction, transmission, modulation and integration events that follow (nociceptive pain). Inflammatory pain derives from damage-induced inflammation to somatic or visceral tissues (actual damage). It can be acute or chronic, depending on the nature of the underlying disease. While the former still has a protective purpose, as it limits movements and further damage until the repair is completed (adaptive pain), the latter lacks any biological purpose (maladaptive pain). Inflammatory pain is the result of nociceptor activation by inflammatory soup mediators released from immune cells, mainly mast cells. This leads to the development of neurogenic inflammation and brings about subsequent neurochemical changes, like wind-up and long-term potentiation, as well as translational and transcriptional modifications (e.g., lower activation threshold of nociceptors and increased expression of functional proteins involved in pain processing). The increased firing rate of the first and projection neuron (i.e., peripheral and central sensitization, respectively) is the main feature of inflammatory pain and leads to primary or secondary hyperalgesia (i.e., increased response to painful stimuli at the site of, or distant to the stimulus) and allodynia (i.e., painful response to harmless stimuli). Neuropathic pain is defined by IASP as “pain that arises as a direct consequence of a lesion or disease affecting the somatosensitive system”. It results from an abnormal activation of the pain pathways, due to a dysfunction or damage to peripheral nerves and/or dorsal nerve roots (peripheral neuropathic pain) as well as spinal cord and/or brain (central neuropathic pain). Accordingly, it is considered disnociceptive, acquires a pathological, maladaptive nature and can be viewed as a disease itself rather than a symptom. Neuropathic pain can last for months to years or possibly even a lifetime, being thus considered a type of chronic pain. Possible mechanisms of peripheral neuropathic pain are (i) persistent hyperexcitability of nociceptors (even after damage repair), (ii) increased excitability of nociceptive fibers following nerve damage (e.g., after dysmyelinosis or neuroma formation), and (iii) structural/functional changes of spinal synapses following nerve degeneration. The resulting burst stimulation of afferent fibers may lead to central sensitization, a hallmark of several painful disorders like feline osteoarthritis [ 55 ]. Central neuropathic pain involves spinal cord and supramedullary neuronal structures and results from lesions affecting the central nervous system or increased activity of thalamic and cortical neurons due to neurochemical changes (e.g., imbalance of glutamatergic/GABAergic transmission). Functional pain occurs spontaneously, in the total absence of tissue damage or evident dysfunction or damage to the nociceptive nervous system. It is probably supported by persistent plastic modifications of the central neuronal circuits induced by nociceptive or dysnociceptive algogenic lesions. As a consequence, originally activated central neuronal circuits remain active even when the lesion has resolved. A possible hypothesis is that mechanisms underlying the spontaneous processing of pain are similar to those that underlie memory: modifications of central neuronal circuits, initially induced by tissue or nerve damage, would remain in the CNS as traces of memory and can be “remembered” even after the lesion has resolved. Functional pain is therefore non-nociceptive, it can last months, years or forever, establishing its chronic nature. It has no biological function and is rather pathological (maladaptive). Like neuropathic pain, functional pain can thus be viewed as a disease itself [ 29 , 54 , 56 ].

As introduced above, chronic pain is an unpleasant experience outlasting the time of healing. Particular cells of the immune system intimately associated with or located within the nervous system, i.e., “non-neuronal cells”, are increasingly acknowledged as major contributors to the development and maintenance of chronic pain [ 51 ]. In particular, mast cells (within the nervous system and in the periphery) and microglia (at spinal and supraspinal level) interact with neurons under physiological and pathological conditions ( Table 1 ). While in the first situation non-neuronal cells support the well-function of neurons (e.g., through releasing neurotrophic factors), in the latter they may become hyper-activated and may cause pain to continue after the original injury has healed [ 94 , 95 ]. In fact, prolonged activation of non-neuronal cells leads to uncontrolled release of pro-inflammatory mediators resulting in long-lasting plastic changes of synaptic connectivity, with enhanced transmission of nociceptive information, alterations of pain signaling pathways and chronic pain development [ 96 , 97 ]. It should also be considered that a bidirectional crosstalk between mast cells and microglia exists [ 98 ] and is currently acknowledged as a critical event in pain hypersensitivity [ 64 , 99 ]. Accordingly, non-neuronal cell hyper-activation—and the resulting neuroinflammation—is a key player of pain states ( Figure 2 ) [ 100 , 101 , 102 ]. Interestingly, non-neuronal cells are also endowed with crucial protective functions in resolution of neuroinflammation and pain [ 59 ]. Indeed, mast cells and microglia are able to reduce sensitization by producing pro-resolution mediators, the so-called specialized pro-resolving lipid mediators [ 103 , 104 , 105 ]. In this framework, particular attention is currently devoted to endocannabinoids and related lipid compounds, such as NAEs and more particularly PEA [ 106 , 107 , 108 , 109 ]. As detailed below, PEA and similar endocannabinoids are locally released on demand during injury to counterbalance the effects of pro-algesic mediators [ 110 , 111 ].

As briefly introduced above, non-neuronal cells not only dangerously boost pain signaling, but also exert crucial functions in resolution of neuroinflammation and pain, through pro-resolution mediators. Among them, endocannabinoids and related NAEs are increasingly being acknowledged to play key roles in pain modulation, with PEA being one of the most studied [ 112 ]. It has been repeatedly found in dozens of vegetable and animal food sources (in nanogram per gram level), from soy to carrots and from eggs to beef [ 39 , 46 ]. Moreover, PEA levels have also been detected in virtually any tissue and body fluid [ 46 , 51 ], where it is enzymatically produced “on demand” in response to actual or potential damage and enzymatically cleaved when it has served its purpose [ 51 , 52 , 113 , 114 , 115 ]. The late Nobel prize winner Rita Levi Montalcini first proposed that PEA acts as an Autacoid Local Injury Antagonist (ALIA), through down-modulating mast cell degranulation [ 116 , 117 ]. It was then found that PEA is synthesized by mast cells and microglia [ 118 , 119 ] and is able to keep cell reactivity within physiological boundaries [ 51 ], thereby controlling neuroinflammation and chronic pain [ 120 , 121 , 122 ]. It has also been demonstrated that PEA not only acts through non-neuronal cells, but may also directly influence neurons. Indeed, PEA was shown to (i) exert protective effects on cultured cortical and cerebellar neurons [ 123 , 124 ], (ii) control spontaneous GABAergic synaptic activity in striatal neurons [ 125 ], (iii) dose-dependently increase intracellular calcium concentration in sensory neurons thereby desensitizing pain receptors [ 126 ]; (iv) modulate the activity of dorsal root ganglion neurons [ 127 ]. On the molecular side, PEA controls neuronal and non-neuronal cells through direct or indirect receptor targets, ranging from canonical to putative cannabinoid receptors, i.e., cannabinoid receptor type 1 and 2 (CB1 and CB2), peroxisome proliferator-activated receptor α (PPARα), transient receptor potential vanilloid 1 (TRPV1) and G protein-coupled receptors 55 and 119 (GPR55, GPR119) [ 65 , 126 , 127 , 128 , 129 , 130 ]. The indirect receptor agonism—i.e., a particular kind of entourage effect—depends on PEA ability to increase the local levels of the endocannabinoids anandamide (AEA) and/or 2-arachydonoylglycerol (2-AG) [ 52 , 125 , 131 , 132 ]. Different types of cannabinoid receptors have been recently localized in canine and feline central and peripheral organs. In particular, the target receptors of PEA have been found in canine and feline skin [ 133 , 134 , 135 ], along the gastrointestinal tract [ 136 , 137 , 138 ], in different brain areas [ 139 , 140 , 141 , 142 ], spinal cord and dorsal root ganglia [ 141 , 143 ]. The distribution of cannabinoid receptors in dogs and cats has been recently addressed by Gugliandolo et al. [ 46 ], to whose paper the reader is referred for more detailed information. The multiple receptor mechanism(s) of PEA is responsible for innate pain control ( Figure 3 ) [ 46 , 52 ] and provides PEA with a natural analgesic function, originally proposed in the late 1990s by Calignano and colleagues [ 144 ] and later even better designed by Piomelli and Sasso [ 145 ]. Currently, the role of body-own PEA in pain control is unquestionably proven by the recent case of a pain-insensitive woman who lacks the NAE degradative enzyme [ 146 ]. She feels almost no pain and has much higher levels of NAEs, with PEA levels being around 4-fold higher than normal. In summary, PEA is an endogenous compound endowed with pain-relieving functions. It is locally produced on demand by non-neuronal cells and other cell types in response to an actual or potential damage, and acts as an endocannabinoid direct or indirect agonist to keep non-neuronal cell response within homeostatic boundaries.

In the last decades, pets are becoming an increasingly important part of family life, being often considered real family members. Owners are more and more often seeking veterinary attention for various diseases affecting their pets, including pain. However, while most information on pain control in dogs and cats exists regarding peri-operative analgesic use, chronic pain conditions are still being undiagnosed and under-treated, especially in the feline species [ 147 ]. Indeed, many conditions may cause maladaptive pain in dogs and cats, as summarized in Table 2 . The incidence of pain in dogs and cats has not received much attention so far. A cross-sectional study on 317 dogs and 112 cats admitted to an emergency service reported that 56% and 54% of dogs and cats respectively were painful, with most dogs suffering from deep somatic pain and most cats from visceral pain [ 149 ]. The percentage was lower in outpatients (1153 dogs and 652 cats), with 20% of dogs and 14% of cats showing evidence of pain [ 150 ]. Neuropathic pain was diagnosed in 7–8% of both species [ 150 , 151 ]. Among the causes listed in Table 2 , one of the most frequent painful conditions in dogs and cats is osteoarthritis (OA), otherwise referred to as osteoarthrosis or degenerative joint disease. The prevalence of canine OA published so far varies widely. In the UK, estimates range from 6.6% in primary-care services [ 152 , 153 ] to 20% based on referral data [ 154 ]. Estimates from North America made on radiographic and clinical data from referral settings show the age-specific prevalence of canine OA, with values ranging from 20% in dogs older than one year to 80% in dogs over eight years [ 155 ]. A cross-sectional study on radiographic signs of feline OA showed an overall prevalence of 92% in randomly selected domestic cats (mean age of 9.9 years) [ 156 , 157 ]. Finally, it should be mentioned that recognizing and measuring pain in animals is anything but easy. Further complicating the issue is the discovery that people rate pain sensitivity differently based on breed-specific stereotypes or phenotypic traits and dog breed archetypes [ 158 ]. Many excellent review papers are available on pain assessment in companion animals, which the reader is referred to [ 22 , 23 , 24 , 159 , 160 ].

As previously discussed, chronic pain—regardless of the underlying cause—may become maladaptive, i.e., without any beneficial role. Neuropathic pain, functional pain and chronic inflammatory pain are all types of maladaptive pain. Any type of maladaptive pain is thus considered pathological and must be treated accordingly. A full discussion on pain management in pets is behind the scope of this article. Briefly, non-steroidal anti-inflammatory drugs (NSAIDs), opioids and steroids alone or associated with adjuvant drugs such as gabapentinoids (gabapentin, pregabalin), NMDA-antagonists (amantadine, memantine), selective serotonin reuptake inhibitors (SSRIs), serotonin-norepinephrine reuptake inhibitors (SNRIs) or tricyclic antidepressants (TCA, e.g., amitriptyline), among others, represent the mainstream pharmacologic treatment of pain [ 27 , 161 , 162 ]. However, when used alone or even in combination, these drugs may still fail to provide complete pain relief [ 149 ]. Moreover, they can lead to the occurrence of adverse effects, especially in the chronic use [ 163 , 164 ]. Chronic pain in pets thus still represents an unmet medical need. The idea that multimodal analgesia tailored to the patient will have most chances of being effective is increasingly being acknowledged in veterinary practice [ 165 , 166 , 167 ]. In this view, dietary intervention with pro-resolving lipid compounds may represent an ideal adjunctive approach. PEA is currently one of the most promising options in this regard.

Before dealing with the effectiveness of PEA in chronic pain, a key formulation question must be addressed. PEA is a highly lipophilic compound and tends to aggregate in large particles (up to 2000 microns)—a big pharmaceutical issue since absorption rate is inversely related to particle size [ 46 , 108 , 168 ]. Particle size reduction through micronization techniques (down to 0.8 microns) importantly improves the dissolution and thus bioavailability ( Figure 4 A) [ 169 ]. This results in superior efficacy of orally administered PEA ( Figure 4 B) [ 170 , 171 , 172 ], while ensuring its safety [ 171 ]. Mainly for this reason, in clinical practice (in which oral route is preferred because of ease of administration) the micronized (PEA-m) and ultra-micronized (PEA-um) forms (collectively known as micro-PEA [ 158 ]) are privileged and are indeed the most investigated. On the contrary, in laboratory animals, the intraperitoneal delivery is generally the easiest and most used administration route. Moreover, it results in faster and more complete absorption compared to oral route [ 173 ]. This is especially true if suspension in carboxymethyl cellulose is used [ 173 ], as it is usually the case with intraperitoneally administered PEA. Indeed, no difference was observed between PEA-um and naïve PEA in pain control, in the event of intraperitoneal delivery ( Figure 4 C), which is absolutely not the case if oral administration is used [ 172 ].

The rationale to administer PEA for pain relief and wellbeing was brilliantly foreseen in the late nineties by the Nobel Prize Winner Rita Levi Montalcini, who stated that “the observed effects of Palmitoylethanolamide appear to reflect the consequences of supplying the tissue with a sufficient quantity of its physiological regulator of cellular homeostasis” [ 117 ]. Since then, several studies in preclinical pain models have been performed, with PEA being given mainly via intraperitoneal route, although intraplantar injection [ 144 , 174 ] and oral administration of micronised formulations [ 175 ] were also used. Interestingly, the concurrent administration of micro-PEA and morphine for 11 days attenuated the development of opioid tolerance [ 176 ], since micro-PEA strengthens morphine analgesia and allows prolonged and effective pain relief with low doses [ 177 ]. Moreover, a descending analgesic mechanism mediated by the serotonergic system has been suggested [ 178 ]. Table 3 and Table 4 summarize the main preclinical investigations. As shown, PEA exerts a clear anti-nociceptive effect in chronic pain models of either inflammatory [ 132 , 144 , 174 , 175 , 176 , 179 , 180 , 181 , 182 , 183 , 184 , 185 , 186 , 187 , 188 , 189 , 190 , 191 ], neuropathic [ 192 , 193 , 194 , 195 , 196 , 197 , 198 ] or mixed nature [ 199 , 200 ]. In particular, it has been found that the anti-nociceptive effect of PEA is comparable to synthetic or plant-derived cannabinoids used for chronic pain, like nabilone [ 181 ] and delta-9-tetrahydrocannabinol (Δ9-THC) ( Figure 5 ) [ 190 ]. Interestingly, an in-press study by Tagne and collaborators has just shown that hemp oil extract with 9.3% cannabidiol by weight has little or no effect when administered alone but synergizes with PEA to produce a greater-than-additive alleviation of neuropathic pain, upon single-dose administration ( Figure 6 ) [ 201 ]. According to the authors, a possible explanation for the observed synergistic interaction lies in the ability of hemp oil extract to improve pharmacokinetic profile of PEA [ 201 ]. Two particularly interesting issues arise from the above findings. First, PEA is not analgesic sensu stricto since it does not modify the physiological pain threshold of control animals and rather electively normalizes conditions of hypersensitivity [ 176 , 185 ]. Second, PEA not only relieves pain itself but also improves pain-induced cognitive impairments [ 198 ]. As far as mechanism of action are concerned, reduced mast cell hyperplasia—even in endoneural sites—and decreased spinal microglia activation were the main observed events [ 120 , 185 ]. At the molecular level, the reduction of markers of pain pathway activation (e.g., Fos) and inflammatory mediators (e.g., cytokines, nerve growth factor) as well as modulation of extracellular signal-regulated kinase (ERK) and nuclear pro-inflammatory factors (e.g., NF-kB) were detected in the spinal cord [ 120 , 132 , 182 , 193 , 195 , 197 ]. Restoration of the glutamatergic synapses homeostasis in the prefrontal cortex and the involvement of de novo neurosteroid synthesis (i.e., allopregnanolone) in the spinal cord were also suggested to mediate PEA-induced analgesia [ 198 , 202 ]. Moreover, electrophysiological signs of decreased neuronal hyper-excitability were reported at the spinal cord level of PEA-um treated neuropathic animals [ 176 , 185 , 200 ]. Finally, the involvement of cannabinoid receptor(s) (e.g., CB2, CB1, PPARα) in the pain-relieving effect of PEA was repeatedly confirmed [ 186 , 192 , 194 , 199 , 202 , 203 ]. The up-regulation of CB2 expression by microglia through PPARα activation has also been suggested as a possible mechanism underlying the pain-relieving effect of PEA [ 204 ]. According to an impressive meta-analysis by IASP Presidential Taskforce on Cannabis and Cannabinoid Analgesia, PPARα agonists and, more specifically, PEA, are effective in attenuating pain-associated behaviors in a broad range of inflammatory or neuropathic pain models [ 205 ]. Pain relieving effect of PEA—mainly given via intraperitoneal route—in animal models of chronic inflammatory pain. Summary of studies in chronological order. Abbreviations. CFA, Complete Freund’s adjuvant; MIA, monosodium iodoacetate; NGF, nerve growth factor; OA, osteoarthritis; PPQ, phenyl-p-quinone. Pain relieving effect of PEA—mainly given via intraperitoneal route—in animal models of neuropathic and mixed pain. Summary of studies in chronological order. Abbreviations. MIA, monosodium iodoacetate; OA, osteoarthritis.

On the clinical side, micro-PEA (i.e., micronized, ultramicronized or co-micronized PEA) has been orally administered as a dietary food for special medical purposes to human patients, either singly [ 206 , 207 , 208 , 209 , 210 , 211 , 212 , 213 , 214 , 215 , 216 , 217 , 218 , 219 , 220 , 221 , 222 , 223 , 224 , 225 ] or in combination with (i.e., add-on dietary intervention to) opioids, gabapentenoids or NSAIDs [ 226 , 227 , 228 , 229 , 230 , 231 , 232 , 233 , 234 , 235 , 236 , 237 , 238 , 239 , 240 , 241 , 242 , 243 , 244 , 245 , 246 , 247 , 248 ], as well as antioxidant compounds (e.g., luteolin, quercetin, polydatin) [ 249 , 250 , 251 , 252 , 253 , 254 , 255 , 256 , 257 , 258 , 259 , 260 , 261 ] in several painful conditions [ 53 , 262 ]. Altogether, nearly 5000 patients have been clinically investigated so far in dozens of published trials, showing an important overall effect in chronic pain, either neuropathic ( Table 5 ), mixed ( Table 6 ) or pelvic pain ( Table 7 ). One of the most interesting findings comes from the neurophysiological assessment of 20 patients with chemotherapy-induced painful neuropathy, with daily administered micro-PEA 300 mg/bid for two months. Besides significant pain reduction, increased conduction velocity of myelinated fibers was recorded, with sensory nerve action potentials from sural and ulnar nerves, compound motor action potentials from peroneal and ulnar nerves and laser-evoked potentials for Aδ fibers being significantly improved [ 212 ]. A further striking finding comes from the so-called “number needed to treat” (NNT), i.e., a measure depicting the effectiveness of an intervention (the lower the NNT, the more effective the intervention). The calculation was elegantly made by researchers from the Department of Human Neurosciences, “Sapienza” University of Rome [ 216 ]. In particular, the percentage of patients who manifested at least 50% pain relief in response to a daily supplementation of micro-PEA 600 mg/die was calculated based on data from a multicenter, double-blind, placebo-controlled, randomized study on 636 patients with low back pain. NNT was found to be 1.7 [ 216 ]. It must be pointed out that it is a remarkable NNT value within the broad panorama of treatments for low back pain in human patients. A systematic review on first-line treatments for neuropathic pain has indeed shown that NNT for 50% pain relief ranges from around 4 to 10 across most positive trials ( Table 8 ) [ 263 ]. The much lower NNT for micro-PEA (i.e., 1.7) emphasizes the good outcome for neuropathic pain relief. The relevance of the data is further strengthened by the non-significant (and indeed infinite) number needed to harm [ 216 ], that is, how many patients must receive a particular treatment for one additional patient to experience a particular adverse outcome. Overall, micro-PEA has shown a very favorable treatment profile in the management of chronic pain in human patients. As far as privately owned animals are concerned, two trials have recently dealt with micro-PEA dietary administration for pain relief. The first is a case series in four jumping horses orally supplemented with PEA-um for non-responsive lameness and significant impairment of athletic performance [ 264 ]. In particular, the diagnoses were the following: navicular syndrome of the left forelimb (1 case), complicated case of chronic navicular syndrome and OA of the distal interphalangeal joint of the right forelimb (1 case), and OA of the distal intertarsal joint of the right hindlimb (2 cases). Horses were fed daily with PEA-um (2.5 g) mixed with a regular mixture of cereals for four months. At the end of the first month of supplementation, lameness on the AAEP scale (American Association of Equine Practitioners 0–5 scale, with zero indicating no perceptible lameness, and five being most extreme) showed improvement in all horses. Three months later, lameness was graded zero, allowing successful return to showjumping without disease recurrence [ 264 ]. The second study is an open-field trial performed in 13 medium-to-large-breed client-owned adult dogs, with chronic OA and persistent lameness longer than one month. All dogs were supplemented for 4 weeks with a complementary feed containing PEA co-ultramicronized with the natural antioxidant quercetin (i.e., PEA-q, 24 mg/kg body weight). The Canine Brief Pain Inventory (CBPI) questionnaire was used to assess the severity of chronic pain (PSS, Pain Severity Score) and how it interfered with the dog’s normal functioning (PIS, Pain Interference Score). With success defined as a reduction of ≥1 in PSS and PIS, treatment was classified as successful in 54.5% dogs as early as week 2 and CBPI scores significantly decreased throughout the study ( Figure 7 ). Moreover, lameness (either scored by the veterinarian on a 0–4 clinical scale or objectively assessed through a dynamic gait analysis) was found to significantly improve during the treatment period [ 265 ]. The findings of the trials summarized above provide clinical evidence on PEA-um (eventually co-micronized with quercetin) as a promising treatment option for chronic pain and related functional disability in horses, as well as dogs.

The revised definition of pain endorsed and approved by the International Association for the Study of Pain (IASP) defines pain as “An unpleasant sensory and emotional experience associated with, or resembling that associated with, actual or potential tissue damage” [ 1 ]. An important change with respect to the previous definition (1979) consists in the recognition that verbally expressing pain is no more a prerequisite to experiencing pain. The IASP further explains that “Verbal description is only one of several behaviors to express pain; inability to communicate does not negate the possibility that a human or a nonhuman animal experiences pain” [ 1 ]. From a veterinary perspective, this represents a definitive recognition of animal pain and poses veterinary practitioners in an “algological position”, i.e., to play a proactive role in recognizing, assessing and managing animal pain. Indeed, many efforts have been made in this direction during the last decades and several European and US groups are moving toward the development of better protocols to detect [ 2 , 3 , 4 , 5 , 6 , 7 ], measure [ 8 , 9 , 10 , 11 , 12 , 13 , 14 ] and treat [ 7 , 15 , 16 , 17 , 18 ] animal pain accordingly. The ever-increasing availability of well-designed pain scales for acute and chronic pain in dogs and cats [ 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 ] and the Pain Management Guidelines [ 27 , 28 ] are good examples. On the treatment side, one of the most up-to-date and clinically relevant issues consists in the multimodal approach to pain management, i.e., a combination of different therapeutic weapons, like analgesic drugs, acupuncture and physiotherapy techniques, as well as dietary interventions [ 29 , 30 , 31 , 32 , 33 ]. With regard to the last measure, calorie restriction and omega-3 fatty acids are the most investigated approaches to chronic pain in pets, particularly osteoarthritis pain [ 34 , 35 ]. Increasing evidence is accumulating on the beneficial effects of N-acylethanolamines (NAEs) in chronic pain. NAEs have been detected in several food sources of vegetable [ 36 , 37 , 38 ] and animal origin [ 39 ]. Moreover, chronic or subchronic high-fat diet, as well as deficient intake of essential fatty acids have been shown to profoundly affect NAE levels in animal body [ 40 , 41 , 42 , 43 , 44 , 45 ]. One of the most studied NAEs is the endocannabinoid-like mediator palmitoylethanolamide (PEA). Its levels in food sources and its pro-homeostatic role have been recently reviewed [ 46 ]. Interestingly, the autoprotective function of PEA was first suggested in dogs, when it was found that the canine myocardium produced PEA in response to ischemic injury [ 47 ] and canine brain possessed the biosynthetic and degradative machinery for PEA [ 48 , 49 ]. Since then, an increasing body of literature has emerged highlighting the importance of dietary intervention with micro-PEA—i.e., the bioavailable form of PEA—for pain relief [ 50 , 51 , 52 , 53 ]. The present paper outlines current information on the involvement of immune cells in chronic pain and reviews the role of endogenous PEA in pain control, as well as the experimental and clinical data on pain relieving effects provided by different PEA formulations. Given that some micro-PEA-containing dietary supplements for dogs and cats are currently being available on the European market, this review wishes to provide scientific evidence to make informed decisions about the management of chronic pain in cats and dogs.

The management of chronic pain is the burden of veterinary practitioners. Multiple pharmacological agents have been employed to treat diverse pathological pain states, including opiates, NSAIDs, anticonvulsants, antidepressants, and others [ 29 ]. However, adverse effects could limit dosing and therapeutic efficacy [ 163 , 164 ]. The recent understanding of the role of non-neuronal cells in pain processing is uncovering potential new targets for managing chronic pain [ 104 ]. Furthermore, it is becoming increasingly clear that enhancing endocannabinoid signalling may prevent patients from developing persistent or chronic pain states mainly through non-neuronal cell modulation [ 266 , 267 , 268 , 269 ]. One such strategy is the dietetic use of the endocannabinoid-like PEA in bioavailable formulations (i.e., micro-PEA). As reviewed here, there is now strong evidence supporting the dietary supplementation with micro-PEA (either as alternative or add-on to conventional treatment) in the management of chronic pain. Such a critical mass of data is being generated that PEA is currently listed among the novel nonopioid interventions to chronic pain [ 270 ]. Although clinical studies in veterinary patients are warranted, the reviewed findings lay the foundations for a scientific and rational use of micro-PEA in the dietary management of chronic pain in dogs and cats.