Neuropathic Pain Assessment Calculator for Dogs, Cats, Horses and Farm Animals

1. Pick the mode. Owner or keeper gives a plain-language result and next steps. Vet or nurse adds the examination state, the conditioned pain modulation test, differentials, syrinx and lateralisation cross-checks, the ChiMPS-T companion scores, medication fields, serial tracking and a printable clinical record. The questions are identical.
2. Choose the species and add the animal's details. Species sets which causes, signs and severity instruments the page offers. Tick any condition on the mimic list (skin, ear, dental, orthopaedic, visceral, behavioural, cardiac, seizures), because each produces the same signs, and tick the 48-hour red-flag box if anything on it has happened; it overrides everything else.
3. Answer the entry screen. How long and in what pattern (more than four weeks or beyond healing, sudden episodes, or acute and still healing), and whether an active injury or disease fully explains the pain. The third criterion, a known lesion or at least one neuropathic signature, is worked out from step 4. Acute pain within healing, or pain fully explained by an active process, exits the neuropathic pathway for now.
4. Grade the pain. Pick the known or suspected cause from the species list, say whether the pain sits where that lesion would put it, tick the sensory findings by type (loss of sensation, heightened sensation, species signature, other pain behaviour), say where the findings are (inside the territory, outside it, or diffuse), record the examination state, say whether a test has confirmed the lesion, and record any drug trial response. The page returns possible, probable or definite with the reasoning, and marks it provisional where the sensory findings are owner-reported, history-only or taken under sedation.
5. Score severity on the species instrument. Every species gets the 0 to 10 pain rating and the 1 to 7 quality of life rating. Dogs add the CBPI for any chronic pain and, for Chiari-like malformation and syringomyelia, the Rutherford seven signs and CHASE. Cats add the FMPI short form. Horses add the headshaking grade and the Horse Chronic Pain Scale. Use the same instrument at every visit.
6. Tick behaviour changes and note sidedness for a dog or cat. The six behaviours are the C-BARQ domains that rose with the Rutherford score; sidedness matched the side of the syrinx on MRI in a later study.
7. Clinician extras. Syrinx width and asymmetry from the MRI report, the ChiMPS-T blocks, and the previous scores and date to see the change.
8. Keep the record. Copy, print, email, download or share a link that restores the exact answers. The blank questionnaire prints the species version for owners to complete at home.

Most pain in animals is nociceptive: an injury, inflammation or disease is stimulating normal pain nerves, and it settles when the cause is treated. The grading system for neuropathic pain assumes you have already decided the pain is not that. Human clinics use screening questionnaires for that step; animals cannot fill them in, so this page uses three criteria drawn from the definitions of chronic and maladaptive pain in the 2022 WSAVA pain guidelines, from the way research studies have enrolled animals with neuropathic pain, and from the exclusion rule in the grading system itself.

1. Duration or pattern. Pain that has persisted beyond the expected healing time or for more than four weeks, or that comes as paroxysmal episodes (sudden bouts, often triggered by touch, movement, excitement or stress) whatever the duration. Ruel and colleagues used exactly this floor, four weeks plus an MRI lesion consistent with the neurological examination, to enrol 29 dogs with naturally occurring neuropathic pain (median duration 12 months). The paroxysmal pattern is characteristic of trigeminal syndromes, feline hyperaesthesia and syringomyelia. The four-week floor stops acute post-surgical pain being graded as neuropathic on day three.
2. A known lesion or at least one neuropathic signature. Either a lesion or disease of the somatosensory system is already known or strongly suspected, or the animal shows at least one of the signs that stand in for the sensations it cannot describe: pain or self-trauma in a nerve, root or dermatome territory; pain from light touch; an exaggerated response to a mild pinch; phantom scratching; skin rippling; self-mutilation; headshaking; stump guarding; pain out of proportion to visible injury. The page works this criterion out from what you enter in step 4.
3. Not fully explained by an active nociceptive process. If an ongoing inflammatory, infectious, orthopaedic, dental, dermatological or visceral process fully explains the pain in that place, treat it and re-screen. If it explains part of the pain, the case is recorded as mixed and continues; both components are treated and each is judged on its own score. Finnerup's grading requires that the pain is not better explained by another diagnosis, and Hunt and colleagues found in 2025 that sensory testing could not pick out a central sensitisation phenotype in osteoarthritic dogs, so orthopaedic pain has to be excluded clinically rather than by the sensory examination.

Failing the screen is not a dead end. The pain is still real; it belongs to the ordinary pain pathway for now, and the screen is repeated at the next visit. Red flags (acute paresis, collapse, unrelenting vocalisation, seizures, signs of aortic thromboembolism in a cat) are checked before the screen and always win.

The International Association for the Study of Pain defines neuropathic pain as pain caused by a lesion or disease of the somatosensory nervous system. The definition says nothing about species. Because animals cannot describe burning, tingling or electric-shock sensations, the diagnosis rests on the lesion and on what can be observed. In people, the 2016 grading system (Finnerup and colleagues) asks three questions and returns a level of certainty; Parker's 2024 review in Frontiers in Veterinary Science adapted it for dogs and cats, and this page applies that adaptation to every species, with the species supplying the list of causes, the list of signs and the severity instrument.

What makes neuropathic pain possible, probable or definite?

1. Possible: a history of a relevant lesion or disease of the nervous system, and pain or pain behaviour in a distribution that lesion would produce (the neck and shoulder for a cervical syrinx; a hind limb for lumbosacral nerve root compression; a stump after amputation or a tail after docking; the tail and back for feline hyperaesthesia; the face for trigeminal-mediated headshaking).
2. Probable: possible, plus either sensory signs in that territory or confirmation of the lesion by a diagnostic test.
3. Definite: possible, plus both sensory signs and a confirmed lesion.

How the sensory signs are weighted

Finnerup states that negative sensory signs (loss of function: reduced or absent sensation, a mappable border) weigh more than positive signs (pain from light touch, an exaggerated response to a mild pinch), because positive signs occur in nociceptive and nociplastic pain too while loss of sensation points to a lesion. This page records both and applies four rules. Loss of sensation in the territory is sufficient on its own for probable. Heightened sensation or a species signature counts for probable only when it is inside the lesion territory (or the territory is not stated); findings recorded outside the territory or on the other side do not count, and the page asks about a second lesion. Diffuse or widespread findings keep the grade at possible, because widespread sensitisation was described in osteoarthritic dogs by Knazovicky and colleagues in 2016 and points to nociplastic or mixed pain rather than a single lesion. Sensory findings that are owner-reported, taken from the history without examination, or obtained under sedation make the grade provisional until a vet has examined the animal unsedated. Ruel's 2020 finding that static sensory thresholds did not separate neuropathic dogs from controls, while the conditioned pain modulation test did, is why clinician mode asks for the cuff test result as supporting evidence rather than for a threshold number.

Confirmation and its ceiling

MRI or CT showing a lesion that explains the distribution, electrodiagnostics, nerve or skin biopsy, or anatomical verification count as confirmation. Finnerup counts surgical verification, so for an amputation, a documented nerve transection, a tail dock, a castration, an onychectomy or a documented thromboembolic episode the procedure or event itself confirms the lesion and the page says so. Three veterinary syndromes are diagnoses of exclusion with no confirmatory test (feline hyperaesthesia syndrome, feline orofacial pain syndrome and trigeminal-mediated headshaking); they can reach probable but never definite. An MRI showing Chiari-like malformation without a syrinx is partial confirmation only, because the malformation is present in nearly every Cavalier. Rusbridge's 2024 review of cats adds that neuropathic pain is suspected when there is abnormal sensation in the distribution of the pain together with a response to a trial of neuropathic drugs, and confirmed by MRI, electrophysiology or a known trauma history; a trial response is therefore recorded here as supporting evidence but does not move the grade, because pain of other kinds also improves on gabapentinoids and single agents often fail in true neuropathic pain. In Ruel's placebo-controlled crossover the owner-chosen outcome improved on placebo as much as on gabapentin while the CBPI pain score and the sensory test did not, which is why the page judges a trial on the validated instrument and never on a global impression.

Causes the page lists

Signs in the screen, by type

Self-mutilation, phantom scratching, skin rippling, frantic tail attack, mouth pawing, headshaking, stamping and stump guarding are counted as species signatures because they are the animal equivalents of paraesthesia and dysaesthesia, the abnormal sensations that define neuropathic pain in people. Weakness, knuckling and a plantigrade stance are motor deficits, and vocalisation, guarding, mood change and reduced lying or feeding are pain behaviours that any pain can produce; they describe the animal and pass the entry screen only when a lesion is known, but they do not raise the grade. The screen is a structured history and examination checklist, not a validated instrument; it exists so the grade is built from stated criteria rather than an impression.

The calculator cannot examine the animal, but it can state what to test and how to record it so that the grade is reproducible between clinicians. The checklist below is adapted from the sensory threshold examination protocol developed in dogs by Sanchis-Mora and colleagues, Moore's 2016 review of canine neuropathic pain, Rusbridge's 2024 feline review and the Ruel 2020 trial.

1. Map the territory first. Decide which dermatome, nerve or region the suspected lesion would affect and mark it on the body diagram in the record. Everything below is tested inside and just outside that border, and compared with the opposite side or a remote site.
2. Light touch and brush. A cotton bud or soft brush stroked across the skin. A pain response to a stimulus that should not hurt is allodynia (heightened sensation).
3. Pinprick or pinch. A blunt pin or haemostat at a level that is mildly uncomfortable in a normal animal. An exaggerated response is hyperalgesia (heightened sensation). A reduced or absent response is hypoalgesia (loss of sensation), the finding that weighs most.
4. Deep pressure and algometer. Record the mechanical threshold where an algometer is available and compare it with the contralateral side.
5. Thermal. Cold latency is the most practical thermal test in dogs.
6. Conditioned pain modulation in dogs (Ruel 2020). Take the algometer threshold at the dorsal metacarpus, inflate a cuff on the opposite forelimb to 200 mmHg for 60 seconds, and re-test immediately. In healthy dogs the threshold rises (an inhibitory profile, mean change +2.3 N); in dogs with MRI-confirmed neuropathic pain it stayed flat or fell (mean -0.2 N, P = 0.045). A facilitatory profile was present in 61.5 percent of neuropathic dogs against 33.3 percent of controls, and gabapentin brought that proportion back to 34.6 percent. Static thresholds did not differ between the groups at all, so the direction of change after conditioning is what the page records.
7. Record where the findings are (inside the mapped territory, outside it, or diffuse) and the state of the animal (unsedated, sedated, or not yet examined). Diffuse findings point to widespread sensitisation and keep the grade at possible; sedated findings are provisional.

Quantitative sensory testing tracks sensory change in dogs with spinal cord injury (Gorney 2016; Hölscher 2025) and the trigeminocervical reflex threshold separates headshaking horses from controls with an area under the curve of 0.83 (Käfer-Karrer 2026), but Hunt and colleagues found in 2025 that QST could not identify a central sensitisation phenotype in osteoarthritic dogs and the 2024 systematic review of QST in canine musculoskeletal pain found limited reliability. Sensory testing therefore informs the sensory step; it never confirms the lesion.

The grade says whether the pain is neuropathic. It says nothing about how bad it is or whether treatment is working; a validated instrument does that, entered once at baseline and again at every visit. No instrument in any species diagnoses neuropathic pain. The table sets out what exists, what each was validated for, and how this page uses it.

The CBPI items, as published

The pain severity score is the mean of the four pain items and the interference score the mean of the six interference items; the quality of life item (poor, fair, good, very good, excellent) is reported separately. The page shows the Ruel 2020 neuropathic cohort values beside the result because they are the only CBPI reference in dogs with confirmed neuropathic pain.

The FMPI short form items, as published

The Chiari-like malformation and syringomyelia evaluation (CHASE) is the first questionnaire for this condition developed with psychometric methods, published by Amanda Valentino, Natasha Olby and colleagues at North Carolina State University with TriviumVet in the Journal of Veterinary Internal Medicine in 2026. Owners are asked to think about their dog's behaviour over the past week and rate how much the dog is scratching, anxious, sensitive, uncomfortable and restless, each from 0 (not at all) to 6 (couldn't be more), giving a total from 0 to 30. A larger candidate list was reduced by factor analysis to two factors, vitality (energetic, lethargic, relaxed, playful) and ease (anxious, sensitive, unhappy, uncomfortable, restless, scratching); the vitality items were dropped because they barely varied in affected dogs even on treatment, and "unhappy" was dropped in piloting because owners did not feel their dogs were unhappy.

What the numbers are

Twenty healthy dogs scored 0 to 6 (mean 1.4, SD 1.8); 29 affected Cavaliers scored 6 to 24 at baseline (mean 15.3, SD 4.8; P < 0.01). Internal consistency was high (Cronbach's alpha 0.89); repeatability was high in healthy dogs (r 0.83 between visits) and moderate in affected dogs between screening and baseline (R² 0.56), which the authors attribute to the fluctuating nature of the disease. In a double-blinded randomised placebo-controlled crossover trial of pregabalin at 5 to 10 mg/kg twice daily (no washout, justified by the 7-hour half-life), CHASE fell on pregabalin compared with both baseline and placebo (P < 0.001 and P < 0.03), and a fall of 4 points or more separated pregabalin from placebo with 63 percent sensitivity and 76 percent specificity (AUC 0.67). Change in CHASE tracked the owner's global impression: dogs rated worse rose by 6.7 points on average and dogs rated better fell by 3.7. Quality of life ratings, quantitative sensory testing and activity monitors did not separate the treatment periods. The authors note that the sensitivity figure assumes pregabalin works in every dog, which it does not, and that part of the placebo improvement is owner expectation.

A follow-up study by the same group (Valentino and Olby, Animals 2026) used a CHASE total of 7 or more to classify Cavaliers as showing clinical signs, and found that those dogs slept worse on the SNoRE 3.0 questionnaire (median 10 versus 7, significant for both sleep quality and dreaming-related interruptions), with a trend that did not survive correction for multiple comparisons towards more snoring and towards sleeping with the head propped up (58 versus 24 percent). That threshold is the one this page uses, and the SNoRE 3.0 sleep score calculator is the companion tool for the sleep component.

How CHASE differs from the Rutherford score

The Rutherford score counts how often seven specific signs occur; CHASE rates how much the dog is affected across five broader descriptors that include mood. CHASE has responsiveness and healthy-dog reference data, which the Rutherford score lacks, so CHASE is the better outcome measure for a treatment trial, while the Rutherford items are the better checklist of what to look for. A dog can be raised on one and not the other, and the page says so when that happens. Neither has been tested in cats; the Rutherford cohort was Cavalier-only, and CHASE's affected dogs were Cavaliers while its healthy reference dogs were non-Cavaliers.

Cats have no validated owner questionnaire for neuropathic pain. The FMPI short form on this page measures musculoskeletal function and the Feline Grimace Scale acute pain; the dog instruments have never been tested in cats. What exists is a recognised list of feline neuropathic pain conditions and a recognised way to reason about them, set out by Clare Rusbridge in a 2024 Journal of Feline Medicine and Surgery review, which is what the cat mode of this page follows.

What does nerve pain look like in a cat?

Spontaneous pain (crying out, sudden aggression, frantic behaviour with no visible trigger), heightened sensitivity to touch (flinching, skin rippling or attacking when stroked along the back or when a collar or harness is put on), and behaviours that suggest abnormal sensation: over-grooming or self-mutilation of one area, tail chasing and tail biting, pawing at the mouth, and exaggerated licking or chewing movements. Cats hide pain by withdrawing, so hiding, reduced jumping, a matted coat from reduced grooming, irritability and litter tray changes are often the first things an owner notices.

Which conditions cause it?

• Feline hyperaesthesia syndrome. Episodes of rippling skin over the lower back, dilated pupils, frantic running, vocalising and attacking the tail or flank, spontaneous or triggered by touch; in a 2019 series of seven cats, most were young males with outdoor access, five had multiple daily episodes and four mutilated the tail. It is a diagnosis of exclusion: the multidisciplinary pathway proposed by Amengual Batle, Rusbridge and colleagues runs dermatology (flea allergy first), neurology (spinal and tail MRI, cerebrospinal fluid, electrodiagnostics) and behavioural medicine in parallel. Gabapentin, alone or combined, produced improvement in six of the seven and remission in five, with recurrence when medication was interrupted.
• Feline orofacial pain syndrome. Exaggerated licking and chewing, pawing at the mouth and self-mutilation of the tongue and lips, often triggered by eating, grooming or mouth movements, exacerbated by stress, and over-represented in Burmese cats. Oral and dental disease, including tooth eruption in young cats, is frequently the trigger and must be treated first; the pain then behaves as neuropathic and responds to anticonvulsant-type agents.
• Diabetic neuropathy. Hind-limb weakness with a plantigrade (walking on the hocks) stance in a diabetic cat; sensory involvement causes pain and self-directed attention to the feet. Glucose control is the treatment.
• Nerve injury after surgery or trauma, including post-onychectomy pain, sciatic nerve injury after femoral or pelvic fracture repair or intramuscular injection, and amputation. Self-mutilation of the site, lameness out of proportion to the healing, and intolerance of touch at the site are the clues.
• Spinal and radicular pain from disc disease, lumbosacral disease, spinal tumours and, rarely, syringomyelia; and trigeminal neuralgia-like facial pain.

How is it treated in cats?

Every drug is off-label in cats and the review recommends starting at the low end of the range and titrating, using more than one agent because single agents have poor success, and treating the cause where one can be found. Ranges cited (Rusbridge 2024): pregabalin 2 to 10 mg/kg orally every 12 hours (preferred gabapentinoid; twice daily is appropriate from its feline half-life); gabapentin 10 to 20 mg/kg every 8 to 12 hours; amantadine 3 to 5 mg/kg every 12 to 24 hours and amitriptyline 0.5 to 4 mg/kg every 12 to 24 hours, each needing at least 3 weeks to judge; topiramate 2.5 to 10 mg/kg every 12 hours; phenobarbital 1 to 3 mg/kg every 12 to 24 hours, which has particular support for orofacial pain; tramadol 1 to 2 mg/kg every 12 to 24 hours as a weak second-line option; ketamine 0.5 mg/kg subcutaneously as a one-off or monthly, or as a 24 to 48 hour infusion in hospital; carbamazepine only when everything else has failed. Licensed NSAIDs reduce the dose of sedating drugs needed. Frunevetmab, the anti-NGF antibody licensed for feline osteoarthritis, is hypothetically useful for peripheral neuropathic pain but unproven. Environmental stress worsens hyperaesthesia and orofacial pain, so the behavioural plan is part of the treatment, not an optional extra. For a fuller owner-level walk-through see neuropathic pain in cats: recognition, diagnosis and multimodal care.

Chiari-like malformation and syringomyelia is the best-characterised neuropathic pain condition in dogs, which is why it has the questionnaires, but it is not the commonest cause. Parker's 2024 review lists degenerative lumbosacral stenosis, a painful radiculopathy, as the other well-recognised canine example, and calls post-stroke pain and painful polyneuropathies under-recognised. In practice the grade on this page is most often used for:

• Intervertebral disc disease and spinal cord injury. Neck or back pain with a lesion on MRI is nociceptive from the disc and neuropathic from the cord and nerve root; touch sensitivity along the spine, self-directed attention to a limb, and pain that persists after decompression are the neuropathic clues.
• Lumbosacral stenosis and nerve root compression. Hind-limb lameness that shifts, pain on tail elevation or lumbosacral pressure, licking at a hind paw or the tail base, and root signature (holding a limb up). Electrodiagnostics (F-waves, cord dorsum potentials) can confirm the root involvement.
• Nerve injury, amputation and surgery. Licking, chewing or mutilating a stump, scar or the distribution of an injured nerve, and intolerance of touch there, are neuropathic signs, not habits, and respond to neuropathic agents rather than to a collar and antibiotics alone.
• Polyneuropathy and diabetic neuropathy. Uncommon in dogs, but a painful polyneuropathy presents as touch aversion of the paws, self-mutilation and weakness.
• Brain lesions. Post-stroke and thalamic lesions can produce central pain that is easy to miss because the animal cannot localise it for you.

How is it treated in dogs?

The same logic as cats: treat the lesion where possible, start a gabapentinoid, add a second agent if the response is partial, and re-examine the sensory signs rather than relying on the owner's global impression alone. Commonly used ranges are gabapentin 10 to 20 mg/kg every 8 to 12 hours, pregabalin 2 to 4 mg/kg every 12 hours as a starting range (the CHASE validation trial used 5 to 10 mg/kg twice daily), amantadine 3 to 5 mg/kg once daily, and amitriptyline 1 to 2 mg/kg every 12 to 24 hours, with NSAIDs for any inflammatory component; verify against your formulary and adjust for renal function and sedation. For Chiari-like malformation specifically, the treatment section of the plain-terms primer below covers the cerebrospinal fluid agents and surgery.

The best-characterised equine neuropathic condition is trigeminal-mediated headshaking: vertical head flicks, snorting, nose rubbing and striking at the face, often worse in sunlight, wind or exercise and often seasonal, attributed to a sensitised trigeminal nerve with no structural lesion on imaging. It is a diagnosis of exclusion: dental, ear, eye, guttural pouch, sinus, nasal, cervical and tack causes are ruled out first, a nose net is a cheap trial, and where available the trigeminocervical reflex threshold is the best objective marker, separating affected horses from controls with an area under the curve of 0.83 in the 2026 Käfer-Karrer study, in which quantitative sensory testing showed no difference. Because there is no confirmatory test the grade on this page cannot exceed probable. Severity is graded with the Talbot scale (0 none, 1 intermittent and rideable, 2 moderate and ridden with difficulty, 3 severe, unrideable or dangerous) or the HRE-S, whose history, rest and exercise subscores showed an inter-observer ICC of 0.98 in the 2024 validation. Treatment evidence is thin: nose nets, cyproheptadine with or without carbamazepine, magnesium, percutaneous electrical nerve stimulation and, in refractory cases, surgical options, all with variable and often temporary results.

Laminitis has a neuropathic component: Jones and colleagues found nerve fibre loss and neuropeptide changes in the digital nerves of chronically laminitic feet in 2007, which is why some chronic laminitic pain responds poorly to anti-inflammatories alone and gabapentin is used as an adjunct. Post-neurectomy and post-trauma neuromas, and cervical, sacroiliac and lumbosacral nerve root disease, are the other recognised causes. Severity in horses outside headshaking is scored on the behaviour section of the Horse Chronic Pain Scale (van Loon and Macri 2021), which has good inter-observer reliability but no cut-off and no hypersensitivity items; the Equine Brief Pain Inventory (Howard 2024) is a promising owner instrument for osteoarthritis whose validation is not yet complete.

No farm species has a validated neuropathic pain instrument, but the evidence that neuropathic pain exists after routine procedures is solid. Sandercock and colleagues characterised traumatic neuroma development after tail docking in piglets in 2016, and Larrondo and colleagues found long-term hyperalgesia at the tail stump and traumatic neuromas in docked lambs in 2019. Castration and dehorning sites carry the same risk. A stump or site that stays sensitive, is guarded, bitten or rubbed long after healing, or that provokes stamping, kicking at the belly or isolation from the group, meets the entry criteria on this page; the documented procedure counts as confirmation of the lesion under Finnerup's surgical verification rule, so an animal with a docked tail, loss of sensation or hyperalgesia at the stump and no other explanation can be graded definite. Severity is tracked with the 0 to 10 rating, a behaviour ethogram (lying, feeding, rumination, stamping, isolation) and, where an algometer or von Frey filament is available, a mechanical threshold at the site. No neuropathic agent is licensed for food-producing species; gabapentin has pharmacokinetic data in sheep, goats and calves but withdrawal periods and food-chain rules apply, so treatment is built on excising a palpable neuroma, licensed NSAIDs, local anaesthesia and a review of the group's procedure technique and analgesia protocol.

Rabbits and small mammals develop neuropathic pain after spinal fracture, disc disease and nerve injury, and self-mutilation of a limb after nerve damage is the classic presentation. Grimace scales are validated for acute pain; for chronic tracking the page uses the 0 to 10 rating and home video, because prey species hide pain in the consulting room. Gabapentin is used empirically alongside meloxicam from exotic formularies. For any other species the grading criteria still apply, and the causes, signs and doses the page shows are generic.

The questionnaire was designed by Lynda Rutherford, Holger Volk and colleagues at the Royal Veterinary College, with Annette Wessmann (Glasgow) and Clare Rusbridge (Stone Lion), and published in The Veterinary Journal in 2012. Its purpose was not to diagnose Chiari-like malformation and syringomyelia but to ask whether the neuropathic pain those conditions cause changes a dog's behaviour and quality of life, as it does in people, where neuropathic pain travels with anxiety, depression and reduced quality of life. The answer was yes: the more often owners saw the seven pain signs, the more fear-related behaviour they reported and the worse they rated their dog's life.

How the questionnaire was built. It was assembled from previously validated instruments: health-related quality of life questionnaires for dogs with heart disease (Freeman et al. 2005) and chronic joint pain (Wiseman-Orr et al. 2006), and the Canine Behavioral Assessment and Research Questionnaire (C-BARQ, Hsu and Serpell 2003) for behaviour. It had three sections. Section 1 collected signalment, general health, confounding conditions, the date of the MRI and current treatment. Section 2 asked how often the owner had observed each of seven neuropathic pain signs on a 5-point semantic differential scale (0 never, 1 seldom, 2 sometimes, 3 usually, 4 always), averaged the seven answers to give the neuropathic pain severity score, used the same scale for a locomotory score and a sleeping and resting score, and asked the owner to rate overall quality of life from 1 (could not be better) to 7 (could not be worse), with 8 for don't know. Section 3 was the C-BARQ, whose eleven behavioural categories are each scored from 0 to 4 and weighted using the published loadings.

Who the dogs were. Owners of Cavaliers with an MRI diagnosis made by a board-certified neurologist at the authors' referral hospitals between January 2004 and March 2010 were invited by letter, and owners were also recruited through breed societies and health forums in the UK, USA and the Netherlands. The web questionnaire was password protected and limited to one response per IP address. Of 564 questionnaires, 303 were excluded as incomplete or incorrectly entered, 201 because the diagnosis was clinical rather than MRI-confirmed, 91 because the MRI section was blank, 11 for no breed, and then dogs with skin disease (21), a heart murmur with coughing or exercise intolerance (31), epilepsy (6), fly-catching (28), otitis media with effusion (33) or another debilitating condition (20) were removed. That left 122 Cavaliers: 95 from the UK, 11 from the USA, 6 from the Netherlands, 4 each from Canada and Australia, and one each from Bermuda and the Czech Republic. Median age was 54 months (range 12 to 132), median weight 8.5 kg (range 5.0 to 13.7), 50 male (29 neutered) and 72 female (42 neutered). Eighty-four dogs (69 percent) were on treatment with one or more of gabapentin, furosemide, pregabalin, prednisolone, carprofen, cimetidine or omeprazole.

What the study found. Spearman rank correlations, corrected for multiple testing with the Benjamini false discovery rate, showed the pain score rising with separation-related behaviour (rho 0.38), non-social fear (0.34), the C-BARQ pain sensitivity domain (0.29), stranger-directed fear (0.28), attachment and attention-seeking (0.24) and excitability (0.21). Aggression did not track the pain score. The pain score also rose with a worse quality of life rating (rho 0.47) and fell with willingness to exercise (rho -0.5) and ability to settle for rest (rho -0.26). Treated dogs had higher pain scores than untreated dogs (median 1.3 versus 0.4) and worse quality of life (median 2, range 1 to 5, versus median 2, range 1 to 3), which the authors read as treated dogs being the more severely affected. The pain score did not correlate with age or weight, and males and females did not differ. Quality of life was worse with increasing weight (rho 0.25).

How the authors interpret it. Fear-based mood change in chronic pain may work by kindling, where each episodic pain event lowers the threshold for the next fear response; the limbic system, amygdala and hippocampus are implicated in both. Dogs that hurt may become more attached and therefore more distressed when left. Reluctance to exercise may reflect fear, touch hypersensitivity or allodynia from collars and harnesses. The conclusion is practical: neurobehavioural changes should be considered in the management of neuropathic pain in Cavaliers with this condition, which means treating anxiety and reduced quality of life as part of the disease, not as a separate behaviour problem.

The sign wording in the table below is reproduced from the paper's methods with attribution. The on-screen prompts are lightly expanded so an owner on a phone knows what each sign looks like; the 0 to 4 scale and the averaging are unchanged, and the expanded wording has not been separately validated.

Each sign is rated on the same scale: 0 = never, 1 = seldom, 2 = sometimes, 3 = usually, 4 = always. The neuropathic pain severity score is the mean of the seven ratings.

How to answer each sign

Answer for how the dog has been recently as a whole, not the single worst day. "Seldom" is a sign you have seen but rarely; "sometimes" means it happens but is not the rule; "usually" means most days or most times the situation arises; "always" means every day or every time. If the dog scratches only when a collar is on or when excited, that still counts, and the pattern is worth telling the vet. Sign 1 asks for scratching that has no skin explanation: a dog with fleas, an ear infection or an allergy will scratch for that reason, and the study excluded those dogs. If your dog has a skin or ear problem, tick it in step 2 and treat it first.

Neuropathic pain score = (sign 1 + sign 2 + sign 3 + sign 4 + sign 5 + sign 6 + sign 7) ÷ 7, reported to two decimal places. The lowest possible score is 0 (no sign ever seen) and the highest is 4 (every sign always). The cohort values below come from the 122 Cavaliers in the study, all with an MRI diagnosis of Chiari-like malformation and syringomyelia (the paper does not separate the two), and 69 percent on treatment. They describe affected dogs, not healthy ones: a score at the cohort median is typical of a dog with the disease, not typical of a healthy Cavalier.

On quality of life, 89 percent of owners rated their dog's life fairly good to excellent (1 to 3), 3 percent (4 of 120) neither good nor bad, and 8 percent (9 of 120) fairly poor. No owner chose poor or could not be worse, so a rating of 6 or 7 is outside anything seen in the study. The authors note that this may partly reflect owners normalising signs or being reluctant to acknowledge suffering in a pet they love, and that owners remain the best source of knowledge about their own dog.

How this calculator bands scores

There is no validated cutoff, so the calculator does not call a dog "painful" or "not painful". It reports where the score sits against the cohort: at or below the cohort median (1), in the upper half of the interquartile range (above 1 up to 1.61), or above the upper quartile (above 1.61), with a further note if it exceeds the highest cohort score (3.43). The untreated group size (38) is derived from 122 minus the 84 treated dogs; the paper reports the treated and untreated medians and ranges but not the untreated n or quartiles. Because treated and untreated dogs scored so differently, it also compares against the treated median (1.3) or the untreated median (0.4) when treatment status is set. Individual signs are flagged as raised at 2 (sometimes) and high at 3 or more (usually or always). These bands are an ASK A VET™ working convention for triage and are not part of the published instrument.

The study's Table 1, reproduced here. Correlations are Spearman rank coefficients between the pain score and the median score for each C-BARQ behavioural category, n = 122. P values are shown with the false discovery rate threshold that applied; a correlation was accepted as significant only where its P value fell below that threshold.

What the numbers mean. These are modest correlations. A rho of 0.38 explains roughly 14 percent of the variation in separation-related behaviour, so the pain score is one contributor among many. The direction is consistent, though: every fear-related and attachment domain moved with pain, and no aggression domain did. The clinical use is not to predict behaviour from the score but to look for these behaviours in a painful dog and to treat them as part of the pain problem. The behaviour checklist in step 6 does not score anything; it records which of the six significant domains the owner reports, so they can be discussed and, if needed, assessed properly with the full C-BARQ.

Quality of life. The scatter of quality of life against the pain score in the paper's Figure 1 shows most dogs clustered at a rating of 2 with pain scores under 1.5, and the higher ratings (4 and 5) almost all sitting at pain scores above 1. In other words, a poor quality of life rating with a low pain score is unusual and deserves a look for another cause; a high pain score with a rating of "good" is common, and may mean the owner has adapted to the dog's signs.

Each sign is characteristic of Chiari-like malformation and syringomyelia pain in Cavaliers, and each has other causes. The study excluded dogs with those other causes, which is why the calculator asks about them. None of the seven signs is specific enough on its own to diagnose the condition; only MRI of the brain and cervical spinal cord can do that.

Sign 1: persistent, compulsive scratching with no skin disease

The classic sign. "Phantom" scratching is a rhythmic scratching action of a hind leg towards the neck, shoulder or ear, usually on one side, often without the paw touching the skin, provoked by excitement, walking on a lead, a collar being touched, or nothing at all. It is associated with a wide syrinx that reaches the superficial dorsal horn of the spinal cord in the C3 to C6 segments on the same side as the scratching leg, and in a 2019 series it was seen only in dogs whose syrinx was 4 mm or wider. Head and ear scratching or rubbing with skin contact is a different thing: it was common in that series but inversely related to syrinx size, and is thought to reflect Chiari-like malformation pain rather than syringomyelia. Rule out: fleas, mites, allergic skin disease, otitis externa, middle ear effusion, anal sac disease (for flank and rump scratching), and a poorly fitting collar.

Sign 2: facial rubbing

Rubbing the face along the ground or furniture, or with the paws, is read as a response to facial or trigeminal discomfort, and headache-like pain is one of the human parallels of Chiari malformation. Rule out: allergic skin disease and pododermatitis (paw licking then face rubbing), otitis, dental pain, nasal disease and foreign bodies, eye pain, and middle ear effusion, which is common in Cavaliers and produces head rubbing, ear scratching and pain on opening the mouth.

Sign 3: hypersensitivity to touch

Allodynia, pain from a stimulus that should not hurt, is a defining feature of neuropathic pain. In these dogs it shows as flinching or crying when the head, ears, neck or shoulders are stroked, when a collar or harness goes on, or when the dog is lifted under the chest. Touch aversion was reported in 30 percent of a 130-dog referral series. Rule out: neck pain from intervertebral disc disease, atlantoaxial instability, steroid-responsive meningitis-arteritis and discospondylitis; ear pain; skin pain from pyoderma or dermatitis; and a fear-based response in an anxious dog, which the study shows travels with pain anyway.

Sign 4: unexplained yelping

Spontaneous vocalisation was the most common sign in the 2019 series (65 percent), typically a sudden, short, high yelp when the dog changes position while lying down, gets up, is lifted, or for no visible reason, and sometimes at night. Crying out when lifted was the most frequently reported pain sign by owners in the Sparks study. Rule out: intervertebral disc disease (the commonest mimic in a spaniel), other spinal pain, abdominal pain, anal sac disease, otitis media, and orthopaedic pain. Yelping with collapse, weakness or inability to rise is an emergency regardless of cause.

Sign 5: reluctance to lift the head

A dog that holds its head low and still, will not look up, or is slow to raise the head for a treat is guarding its neck. Cervical pain from any cause does this: Chiari-like malformation, syringomyelia, disc disease, atlantoaxial instability, meningitis, discospondylitis, trauma and neck masses. In Cavaliers specifically, sleeping with the head propped up or a preferred head posture during sleep was one of the strongest owner-reported predictors of a large syrinx in a 2024 study.

Sign 6: reluctance to bend the neck to eat

The same neck-guarding seen at the food bowl: standing over the bowl, eating only from a raised bowl or a hand, or lying down to eat. The differentials are the cervical ones listed for sign 5, plus dental and oral pain, pharyngeal disease, and nausea, which produces a different picture (interest then turning away, lip licking, drooling).

Sign 7: reluctance or pain when defaecating

Posturing to defaecate loads the lumbosacral spine and the abdominal wall; dogs with thoracolumbar syringomyelia or spinal pain hesitate, circle, or cry while straining. Rule out: constipation, colitis and anal sac disease (strain, frequency, blood or mucus), perineal hernia, prostatic disease in males, hip or stifle pain, and lumbosacral disease. This sign is also the one most easily confused with a bowel problem, so faecal consistency and frequency belong in the history.

Patterns

Scratching, rubbing and touch hypersensitivity together point at the head and neck. Signs 5 and 6 together mean cervical pain until proven otherwise, and in a non-Cavalier that is a disc, atlantoaxial or meningitic work-up rather than a Chiari one. Sign 7 alone, without the others, is more often a bowel, anal sac or lumbosacral problem. Yelping with no other sign and a normal neurological examination still needs a spinal cause excluded.

Six years after Rutherford, Courtney Sparks, Natasha Olby and colleagues at North Carolina State University published a second owner tool, the Chiari-like malformation pain and scratch tool (ChiMPS-T), alongside body maps (ChiMPS-M) on which owners shade the areas their dog scratches or seems to find painful (Sparks et al., J Vet Intern Med 2018). Its aim was a tool for clinical trials, and it separates scratching from pain, which the Rutherford score averages together. The scores are built as follows.

What the study found. Fifty client-owned Cavaliers aged 15 months to 11 years, recruited by owner volunteering, had the questionnaire, the maps, a neurological examination and MRI. Thirty had syringomyelia. Owners reported signs in 33 dogs (scratching in 32, pain in 23), and the commonest pain sign was crying out when lifted (11 dogs), then head shyness (6), side pain (5), running from pain (4), neck pain (4), and exercise-induced pain, being woken by pain and collar sensitivity (1 each). Only 34 owners completed the detailed score version. Median scratch score was 2.5 (range 0 to 14.5) in dogs without syringomyelia and 4 (0 to 11.5) with it; median pain score was 0 (0 to 10) without and 2 (0 to 12) with. Neither difference was significant, and neither score, nor the shaded map area, nor the presence of owner-reported pain or scratching, was associated with the presence of a syrinx, its maximum height, or pain on neurological examination. The one strong association was sidedness: owners reported lateralised signs in 10 of 15 dogs with an asymmetric syrinx and in only 3 of 35 without (P < 0.0001). Ten dogs with Chiari-like malformation but no syrinx showed classic neuropathic pain signs.

How this page uses it. In clinician mode the ChiMPS-T blocks are optional. Each total is computed only when its components are complete, the working is shown, and the result is compared with the study medians. The published checkbox list for the signs-of-pain item sits in the paper's supplementary questionnaire, which is not reproduced here; the eight signs on this page are the categories reported in the paper's Figure 1 across all 50 dogs, so the checklist is a reconstruction. Because the published score allows 0 to 6 sign-of-pain points, the calculator caps that component at 6 for the total and states the raw count alongside. These scores have not been tested for reliability or responsiveness, and the authors say so; use them to describe, not to decide.

Chiari-like malformation (CM) is a mismatch between the size of the back of the skull and the brain inside it. The cerebellum and brainstem are crowded into and through the opening at the base of the skull (the foramen magnum), which obstructs the normal flow of cerebrospinal fluid between the head and the spinal canal. Almost all Cavaliers have it; the 2012 paper cites an estimated 95 percent. Syringomyelia (SM) is the fluid-filled cavity, a syrinx, that can then form inside the spinal cord, most often in the neck, as pressure waves in the cerebrospinal fluid are transmitted into the cord. Around 70 percent of Cavaliers develop it, and prevalence rises with age: a 2011 screening study found syringomyelia in about a quarter of one-year-olds and about 70 percent of dogs over six years.

Why it hurts. The syrinx damages the dorsal horn of the spinal cord, the grey matter that processes sensory input. Damaged dorsal horn neurons reorganise and become spontaneously active, producing the four things neuropathic pain does: dysaesthesia (unpleasant abnormal sensations, which is what phantom scratching probably represents), allodynia (pain from touch that should not hurt), hyperpathia (exaggerated pain from a painful stimulus) and spontaneous burning pain. The crowding at the foramen magnum is thought to add headache-like pain by compressing the trigeminal nuclei in the brainstem, which is the leading explanation for pain in dogs that have the malformation but no syrinx. Syrinx width predicts pain better than anything else on the scan: in a 2007 study 95 percent of dogs with a maximum syrinx width above 6.4 mm showed pain, and the 2019 series found phantom scratching, scoliosis, weakness and postural deficits only in dogs with a syrinx 4 mm or wider. The degree of cerebellar herniation does not predict pain.

Which dogs. The Cavalier is the breed the condition is defined in, but it is also recognised in the King Charles Spaniel, Griffon Bruxellois, Chihuahua, Pomeranian, Yorkshire Terrier, Affenpinscher, Maltese, Papillon, French Bulldog and Boston Terrier, among others. The score on this page was developed and its cohort values measured in Cavaliers only.

Diagnosis. Only MRI of the brain and cervical spinal cord confirms the malformation and measures the syrinx; radiographs cannot see it, and a neurological examination is often normal. Because syringomyelia is so common in Cavaliers without signs, a syrinx on MRI does not prove the signs come from it, and the 2019 series lists the conditions that mimic or coexist: intervertebral disc disease, otitis media with effusion, idiopathic epilepsy, fly-catching, degenerative myelopathy, and skin disease. A dog that scratches, yelps and guards its neck needs those excluded even when the MRI shows a syrinx.

Treatment. Medical treatment aims at the neuropathic pain and, less certainly, at reducing cerebrospinal fluid pressure. Gabapentin and pregabalin are the usual first line for neuropathic pain; a randomised controlled trial of pregabalin in Cavaliers with syringomyelia-related pain published in 2020 found it reduced owner-assessed pain compared with placebo. Non-steroidal anti-inflammatories help the inflammatory and musculoskeletal component. Omeprazole or cimetidine, furosemide and, for refractory cases, corticosteroids are used to reduce cerebrospinal fluid production, on weaker evidence. Amantadine, amitriptyline, tramadol and acupuncture are used as adjuncts. Foramen magnum decompression surgery relieves the obstruction and improves most dogs in the short term, but signs commonly recur and many dogs still need medication, so surgery is generally reserved for dogs that fail medical management or are deteriorating. Weight control, a harness rather than a collar, raised food bowls, soft bedding and avoiding pressure on the neck are the everyday measures. Every dog in the 2012 cohort that was on treatment was still showing signs, and treated dogs had the higher scores: treatment reduces signs, it rarely removes them.

Breeding. The British Veterinary Association and Kennel Club run an MRI screening scheme for Cavaliers, grading Chiari-like malformation and syringomyelia so that breeders can select against them. Ask a Cavalier breeder for the MRI grades of both parents.

Rusbridge, McFadyen and Knower 2019. A retrospective series of 130 Cavaliers seen by one neurologist with full neuroaxis MRI, grouped by maximum syrinx transverse diameter (none; 0.5 to 1.99 mm; 2 to 3.9 mm; 4 mm or more). The commonest signs were vocalisation (65.4 percent), spinal pain (54.6 percent), reduced activity (37.7 percent), reduced ability with stairs or jumping (35.4 percent), touch aversion (30.0 percent), altered emotional state (28.5 percent) and sleep disturbance (22 percent). Head scratching or rubbing (28.5 percent) was inversely associated with syrinx size and not associated with middle ear effusion, while phantom scratching, scoliosis, weakness and postural deficits were seen only in the 4 mm or more group. The authors conclude that signs of pain are common but not syrinx-dependent, which supports the idea that the malformation itself causes pain, and that a wide syrinx involving the dorsal horn produces phantom scratching on the same side and torticollis with the shoulder deviated to the same side and the head tilted to the other.

The 2024 Acta Veterinaria Scandinavica study. An owner questionnaire of 14 yes-or-no, one sidedness and five 4-point frequency questions in Cavaliers with MRI, aimed at predicting a large syrinx. A preferred head posture during sleep had a positive predictive value of 0.82 for a syrinx of 4 mm or more; unprovoked, neck or phantom scratching 0.73; and the combination of scratching, aversion to touch and a preferred sleeping head posture 0.91. These are the owner observations most worth asking about when deciding whether to image.

CHASE (2026). The first psychometrically developed instrument for the condition, scored in step 5 and described in its own section. Its five descriptors overlap closely with the Rutherford signs and the behaviour domains on this page, and it is the only one with responsiveness data.

Sleep (Valentino and Olby, Animals 2026). Thirty-seven Cavaliers scored with CHASE and SNoRE 3.0: dogs with clinical signs (CHASE 7 or more) had significantly worse sleep quality and more dreaming-related interruptions, and a non-significant trend towards more snoring and sleeping with the head propped up, which the authors hypothesise reflects brainstem compression and avoidance of the rise in intracranial pressure that comes with lying flat. Sleep is therefore a legitimate part of the assessment, and the difficulty settling item in step 5 and the SNoRE 3.0 calculator cover it.

Clinician tools. Quantitative sensory testing and the nociceptive withdrawal reflex have been studied as objective measures of neuropathic pain in Cavaliers, and a specialist neurological examination records neck and back pain on palpation, inducible scratching, postural deficits and scoliosis. None replaces the owner's account, because the signs are episodic and often absent in the consulting room; seven of the 33 symptomatic dogs in the Sparks study were not painful on examination.

• It is owner interpretation. Measuring neuropathic pain is hard even in people, who can describe it, and the whole study rests on owners' observations. The authors reduced bias by asking about observable signs rather than asking owners to rate their dog's pain, but response bias remains.
• The criteria are inherited. The seven signs come from earlier clinical descriptions (Rusbridge et al. 2007), not from a psychometric development process, and the score has no reported reliability, internal consistency or responsiveness data. That is why this page does not offer a minimal important change.
• Selection. Only 122 of 564 responses survived; respondents were owners engaged enough to find the study through a clinic letter or a breed forum, and 78 percent of the final cohort came from the UK. Treated dogs, which were 69 percent of the cohort, were more severely affected.
• Weight was owner-reported, not body condition scored, so the weight and quality of life finding is approximate.
• Owners may under-rate suffering. Only 8 percent judged their dog's quality of life poor and 3 percent indifferent, which the authors suggest may reflect cognitive dissonance in owners who love a dog they cannot fully help, and the breed's naturally affectionate temperament masking the impact of pain.
• Correlations, not causes. A rise in fear-related behaviour with the pain score does not prove the pain causes the behaviour; the authors propose kindling and limbic sensitisation as a mechanism but did not test it.

Every score on this page is worth more as a trend than as a single value. For the Rutherford score: score before starting or changing treatment, again at 2 and 4 weeks, and at every recheck; a two-week gap is a reasonable minimum because gabapentin and pregabalin are usually titrated over that time and the owner needs enough days to judge frequency. Keep the same person answering: frequency ratings of "sometimes" versus "usually" differ a lot between household members, and no study has measured agreement between two owners of the same dog on this questionnaire. Because no minimal clinically important difference has been published, the calculator reports the raw change and treats a shift of one step on a single sign (about 0.14 on the average) as noise. As an ASK A VET™ working convention it flags a rise of 0.5 or more as worth a look, and it never claims a fall proves the treatment worked, because the signs are episodic and wax and wane with weather, excitement and handling. Record quality of life at the same time; a falling pain score with a static or worsening quality of life rating means something else is going on.

What phantom scratching looks like

The dog stops, often mid-walk or when a collar is touched, and one hind leg scratches rapidly towards the neck or shoulder without the paw reaching the skin. It is usually the same side each time and often stops if you distract the dog. Ordinary itch scratching makes contact and leaves marks; this does not. Both can happen in the same dog, which is why the skin and ears must be checked first.

What to film

A phone video of a scratching episode, a yelp when your dog is lifted or gets up, the way it eats from a bowl on the floor, and how it holds its head at rest is worth more to your vet than any description. Film in good light, from the side, and include the dog's whole body. Note the time of day, what was happening and whether the collar was on.

When to worry

Today: sudden weakness or wobbliness, unable to stand or walk, a limb that drags or knuckles, screaming that will not settle, a seizure, or a new head tilt or twist in the neck. This week: yelping or touch sensitivity that is now "usually" or "always", a dog that has stopped eating from its bowl, a quality of life rating of 5 or worse, or a score that has climbed since you last did it. Next routine visit: a stable low score with occasional signs, especially if the dog is already on treatment and comfortable.

What to bring to the vet

This record (Copy or Print it), your videos, the MRI report if the dog has had one, the names and doses of everything the dog takes, and a note of what makes the signs better or worse: collar, excitement, cold weather, being lifted, mornings.

Everyday measures that help

Use a well-fitted harness rather than a collar, raise the food and water bowls, provide a soft bed that supports the neck, lift the dog with one arm under the chest and one under the rump rather than under the armpits, keep weight lean, and keep exercise regular but gentle. None of these replaces pain medication in a dog that needs it.

If your dog is not a Cavalier

The signs still mean something, but the comparison numbers on this page do not apply, and in most other breeds neck pain, scratching and yelping are more likely to come from a slipped disc, an ear problem or skin disease than from Chiari-like malformation. Your vet will decide whether MRI is the right next step.

Neuropathic pain (NeP)

Pain arising as a direct consequence of a lesion or disease affecting the somatosensory system, the nerves and spinal cord pathways that carry sensation. It is chronic, poorly responsive to ordinary painkillers, and produces abnormal sensations as well as pain.

Chiari-like malformation (CM)

A developmental mismatch between the back of the skull and the brain in which the cerebellum and brainstem are crowded into the foramen magnum, obstructing cerebrospinal fluid flow. Present in nearly all Cavalier King Charles Spaniels.

Syringomyelia (SM)

A fluid-filled cavity (syrinx) within the spinal cord, usually in the neck, that forms when cerebrospinal fluid flow is obstructed. It damages the dorsal horn and causes neuropathic pain, phantom scratching and, when wide, weakness and scoliosis.

Phantom scratching

Rhythmic scratching of a hind leg towards the neck or shoulder without the paw making contact with the skin, associated with a wide syrinx involving the dorsal horn on the same side.

Allodynia

Pain from a stimulus that does not normally cause pain, such as light touch, brushing, a collar or tack. Sign 3 of the Rutherford score and a positive sensory sign in the grade.

Dysaesthesia

An unpleasant abnormal sensation, spontaneous or evoked; the burning, pins and needles or crawling sensations people with neuropathic pain describe, and the likely basis of phantom scratching.

Hyperpathia

An exaggerated pain response to a stimulus that is normally painful, often with the pain outlasting the stimulus.

Dorsal horn

The sensory grey matter of the spinal cord where incoming pain and touch signals are processed. Damage here is what turns a syrinx into pain.

C-BARQ

The Canine Behavioral Assessment and Research Questionnaire, a validated owner questionnaire that scores behaviour in eleven categories including stranger-directed fear, non-social fear, separation-related behaviour, attachment and excitability. Section 3 of the Rutherford questionnaire.

Semantic differential scale

A rating scale anchored by opposite words, here never to always, scored 0 to 4.

Spearman rank correlation (rho)

A measure from -1 to 1 of how consistently two rankings move together. 0.3 is a modest association, 0.5 moderate.

False discovery rate (FDR)

A correction for testing many correlations at once, which sets a stricter P value threshold so that chance findings are not reported as real.

Otitis media with effusion (PSOM)

Fluid in the middle ear behind an intact eardrum, common in Cavaliers, which causes head rubbing, ear scratching and pain and was an exclusion in the study.

ChiMPS-T

The Chiari-like malformation pain and scratch tool of Sparks et al. 2018, which separates a scratch score (0 to 16) from a pain score (0 to 20).

Neuropathic pain grading

The possible, probable and definite levels of certainty from the 2016 IASP grading system, adapted to animals by Parker in 2024: relevant lesion plus matching distribution gives possible; add sensory signs in that territory or a confirmed lesion for probable; both for definite.

CHASE

Chiari-like malformation and syringomyelia evaluation: five words (scratching, anxious, sensitive, uncomfortable, restless) rated 0 to 6 for the past week, total 0 to 30; 7 or more marks clinical signs and a fall of 4 or more identified pregabalin responders in the validation trial.

Feline hyperaesthesia syndrome

Episodes of rippling skin over the lower back, frantic running, vocalising and attacking the tail in cats, treated as neuropathic pain once skin, spinal and behavioural causes are excluded.

Feline orofacial pain syndrome

Exaggerated licking, chewing and pawing at the mouth with self-mutilation of the tongue or lips, often triggered by eating and worsened by stress, over-represented in Burmese cats.

Paraesthesia

An abnormal sensation that is not unpleasant, such as tingling; the likely basis of phantom scratching and skin rippling. When unpleasant it is called dysaesthesia.

Entry criteria

The three ASK A VET™ screening questions applied before grading: pain beyond four weeks or normal healing or in sudden episodes; a known lesion or at least one neuropathic signature; and pain not fully explained by an active injury or disease.

Hyperalgesia

An exaggerated pain response to a stimulus that is normally mildly painful, such as a gentle pinch. A positive sensory sign; counts toward the grade when it is inside the lesion territory.

Hypoaesthesia and hypoalgesia

Reduced sensation, and reduced response to a painful stimulus, in an area. The negative sensory signs, which weigh most in the grade because loss of function points to a lesion.

Nociplastic pain

Pain from altered pain processing without evidence of tissue damage or a lesion of the somatosensory system; the likely explanation when sensory findings are diffuse rather than confined to a lesion territory.

Mixed pain

Pain with both a nociceptive and a neuropathic component, recorded when an active process partly explains the pain; each component is treated and scored separately.

Conditioned pain modulation

The reduction in pain sensitivity that a second painful stimulus normally produces; tested in dogs with a limb cuff and an algometer. A flat or falling threshold after conditioning (a facilitatory profile) was more common in dogs with neuropathic pain in Ruel 2020.

CBPI

The Canine Brief Pain Inventory: four pain severity items and six interference items scored 0 to 10, plus a quality of life item, validated for chronic pain in dogs and used in dogs with confirmed neuropathic pain.

FMPI-sf

The short-form Feline Musculoskeletal Pain Index: nine everyday activities rated 0 (normal) to 4 (impossible), validated for degenerative joint disease in cats.

Trigeminal-mediated headshaking

Neuropathic facial pain in horses from a sensitised trigeminal nerve, producing vertical head flicks, snorting and nose rubbing; a diagnosis of exclusion graded 0 to 3 for severity.

Traumatic neuroma

A disorganised regrowth of nerve fibres at the cut end of a nerve, found after tail docking, castration, amputation and neurectomy, and a source of persistent hyperalgesia.

Diagnosis of exclusion

A condition diagnosed by ruling out the alternatives because no test confirms it (feline hyperaesthesia, feline orofacial pain, trigeminal-mediated headshaking); on this page such a cause can reach probable but never definite.

Working convention

A threshold or band set by ASK A VET™ for triage on this page, labelled as such because the published score has no validated cutoff.