Chiari-like malformation and syringomyelia

Syringomyelia (SM) is defined as an abnormal accumulation of fluid within the parenchyma of the spinal cord, leading to the formation of a cavity known as a syrinx (Figure 1). SM is thought to arise primarily due to abnormalities in the flow of cerebrospinal fluid (CSF), and several conditions have been suggested as causative factors for SM in dogs, Chiari-like malformation (CM) being the most common 1. CM and SM may occur independently or concurrently (CM/SM) within affected individuals, and may manifest symptomatically or asymptomatically.

In dogs, CM is characterized by the presence of an abnormally shaped supraoccipital bone, leading to rostral compression of the caudal aspect of the cerebellum (Figure 2). There is a mismatch between the structures of the caudal cranial fossa, causing the cerebellum to herniate into the foramen magnum 2 3. It is notable that the features of canine CM are analogous to Chiari malformation type 1 (CM1) in humans 4, where it can either be a congenital condition or acquired if the proportions of the cranial cavity are altered in some way. Between 70-80% of humans with CM1 have accompanying SM.

Pathophysiological mechanisms

Anatomical changes

Some toy breeds, such as the Cavalier King Charles Spaniel (CKCS), have a smaller caudal cranial fossa and abnormal occipital bone compared to mesaticephalic dogs 2. They are also 80% more likely to show closure of the spheno-occipital synchondrosis at an earlier age than other brachycephalic breeds or mesaticephalic dogs 5, which decreases the ability of individuals to accommodate changes in volume of the hindbrain 3 4.

As with humans with CM1, affected dogs have an increased volume of brain parenchyma in comparison to the volume of the caudal cranial fossa 6 7. In addition, it is known that the larger the foramen magnum, the greater the risk of cerebellar herniation 8 9.

Fluid dynamics

Normally, CSF circulates from the ventricular system to the subarachnoid space. A decrease in venous drainage due to a reduced volume of the venous sinuses and narrow jugular foramina could in turn lead to a reduction in CSF absorption 10. Movement of CSF within the subarachnoid space of the cranial cavity and into the spinal cord is dependent on cardiac systole and intracranial arterial pulsations. The syrinx development seen with SM is due to obstruction of the normal CSF flow and a suction (Venturi) effect caused by a reduction in the width of the subarachnoid space in the region of the cervical segment between C1 and C3 11. These forces are due to the differential pressure during pulsations, which are significantly greater in CKCS with SM 12. One study reported that turbulent CSF flow at the foramen magnum was associated with SM severity and that the velocity of the CSF dorsal to C2/C3 was inversely related to the presence of SM 13.

Prevalence and genetics

CM is encountered most commonly in the CKCS breed; with up to 92% of the population affected 9, although this includes both silent and symptomatic dogs. It is also prevalent in the Griffon Bruxellois 14 and many other small breed dogs, such as the French Bulldog, Chihuahua, Pomeranian, Maltese Terrier, Pug and Yorkshire Terrier.

One study reported that asymptomatic SM was found in 25% of CKCS up to the age of 12 months old, with 70% of the dogs developing SM (symptomatic or asymptomatic) by the age of 6 years of age 15. Another study reviewed the clinical status of 54 dogs older than 5 years with known CM/SM: the mean follow-up time was 71 months after the initial diagnosis, and 32% of the dogs which had been initially asymptomatic were found to be symptomatic at re-evaluation 16.

A recent report found the prevalence of symptomatic CM/ SM in the CKCS breed diagnosed within primary care UK practices to be 1.6% 17, although the authors acknowledged that this figure is likely to be an underestimation of the real numbers affected. Other breeds diagnosed with symptomatic CM/SM included the King Charles Spaniel, Affenpinscher, Chihuahua and Pomeranian.

The heritability of SM has been estimated in CKCS as moderately high. Recently, a study identified that mixed breed dogs are at lower risk of CM/SM 18, but careful selection of appropriate morphologic characteristics is still needed to reduce the degree of the abnormally shaped supraoccipital bone. In fact, some CKCS-cross breed dogs have been diagnosed with symptomatic CM/SM by magnetic resonance imaging (MRI) 17.

Quantitative genetic trait loci (QTL) analysis of CM/SM has been undertaken in the Griffon Bruxellois which identified six highly significant traits for CM/SM 19. These are associated with Canis Familiaris Autosome 2, which is strongly linked with the height of the cranial fossa and contains a single candidate gene, Sall-1, that has previously been linked to CM1 in humans. While genetic investigations are necessary both to understand the heritability of the condition and to attempt to breed out the malformation from affected lines, the improvement of welfare, understanding of the impact on quality of life of CM/SM dogs and owners, and the design of better veterinary treatment should be the priorities for palliative actions until better breeding strategies can be implemented.

Clinical signs

In humans with CM1, reported clinical signs include pain, headaches, altered sensation, weakness, dysphagia, sleep apnea, sensory deficits, weakness of the extremities, and muscle atrophy. The pain experienced with SM has been characterized as central neuropathic pain (NeP). NeP is defined as “pain caused by a lesion or disease of the somatosensory nervous system” and central NeP can result from any type of injury to the central nervous system. In the case of SM, NeP is caused by direct damage to the dorsal horn and spinothalamic tract and in human patients is described variously as burning or cold sensations, prickling, tingling, pins-and-needles, stabbing, shooting, tight, swollen, and squeezing sensations. This is important to remember, as our canine patients may suffer all of these NeP signs, although we cannot always recognize them.

Dogs of all ages may present with symptomatic CM. However, the most common signs develop in younger dogs at two to four years of age, with a range from a few months to older than 10 years of age 8 20. Symptomatic dogs can also present with CM without SM. The most common signs reported with CM/SM are manifestations of generalized pain (Figure 3) or pain localized to the spinal cord (typically the neck) including spontaneous vocalization, sensitivity to palpation, and avoidance of pain-evoking movements/postures 11 17. Owners may also notice behavioral changes such as increased anxiety. Other signs commonly exhibited by dogs clinically affected with CM/SM include “air guitar/phantom scratching” (scratching without making contact with the skin) (Figure 4) which may be associated with dysesthesia (abnormal sensation) or itch. Neuropathic pain can manifest itself through allodynia (pain from a stimulus that is not normally painful), hyperalgesia (increased pain sensation from a painful stimulus), and paresthesia (spontaneous pinprick sensation).

As with the situation in humans, dogs with SM are thought to exhibit pain secondary to disruption of the fibers synapsing on the dorsal horn laminae. Asymmetric syrinxes have been shown to cause NeP, with syrinx width (as assessed by MRI measurement) being the strongest predictor of pain, scratching and scoliosis in one study 21. In the same study, concentrations of substance P, an excitatory neurotransmitter, and interleukin-6 were significantly elevated in the CSF of dogs with NeP and SM compared to asymptomatic dogs. The release of these substances may initiate pro-nociceptive effects, leading to development of NeP and continuation of the sensitization and activation of the receptors involved.

Diagnosis

Because of the overlapping nature and low specificity of clinical signs that can be confounded with other diseases (e.g., dermatological diseases causing scratching or otitis externa), MRI is the diagnostic of choice. Reaching a definitive CM/SM diagnosis can prove challenging, especially in cases where MRI is unavailable or not affordable by the owner. A recent study found that dogs are more likely to be diagnosed — and therefore treated accordingly — if they are insured 17. However, even if a final diagnosis cannot be reached due to financial limitations, it is important for veterinary practitioners to recognize pain in these patients and treat appropriately, especially as the pain is chronic in nature.

History and clinical signs aid in the diagnosis, especially for the presence of NeP. To meet the diagnostic criteria for NeP in humans, clinicians should evaluate:

1. the neuroanatomical distribution (i.e., the spinal cord)
2. if the patient’s history relates the pain to a relevant lesion (reluctance to being touched on the neck related to the location of the syrinx)
3. the demonstration of a lesion (i.e., detection of a syrinx) on the MRI scan and also demonstration of somato-sensory disturbances concordant with the distribution of the pain (allodynia/hyperalgesia on the neck through quantitative sensory testing).

A similar approach could potentially be used for affected dogs, using quantitative sensory testing (QST) to assess for NeP. This relies on precision instruments that apply a stimulus to an area of the body to quantify sensory or nociceptive thresholds (Figure 5). Touch and vibration detection, as well as mechanical and thermal (heat and cold) detection of noxious stimuli are the tests employed in both humans and veterinary patients. Currently, QST is only used in the veterinary sphere as a research tool to prove the presence of central sensitization and hyperalgesia in dogs with chronic pain 22, although its use in the clinical setting could be very useful in the near future.

Chronic pain scales are very useful for diagnosis of possible chronic NeP and to evaluate changes over time and the effects of treatment. The Canine Brief Pain Inventory* may be useful, although so far it has only been validated in dogs for osteoarthritis and cancer.

www.vet.upenn.edu/research/clinical-trials/vcic/pennchart/cbpi-tool *

Treatment

As with the human situation, management possibilities for CM/SM include surgical or medical options. Surgical intervention aims at restoring some normality to CSF dynamics via decompression of the caudal foramen magnum by removing part of the supraoccipital bone. One paper that surveyed 15 CKCS dogs reported 80% of patients improved post-surgery, whilst 20% showed no improvement 20. The clinical outcomes were monitored for more than 12 months postoperatively, but the author noted that the procedure seemingly does not result in syrinx collapse and resolution, and any clinical improvement may not be permanent, with some dogs deteriorating after an initial response.

Medical treatment of canine CM/SM utilizes three main approaches: analgesia, reduction of CSF production, and empirical use of anti-inflammatories such as glucocorticoids. This review will focus on pain relief. The objective is to improve the patient’s quality of life (QoL), usually via a combination of two or more drugs, as current management options available will not completely resolve the pain. Owners must be advised that this type of pain will probably persist throughout the entire life of their pet and that it may be challenging to treat successfully. One report noted that only 13% of symptomatic dogs with CM/SM had improved with medical management, whereas 56% had worsened despite treatment 16. Owners should therefore have clear expectations of the treatment goals and a pain scoring system that the owner can use conveniently at home, such as a Visual Analogue Scale (VAS), should be adopted for ongoing assessment (Box 1).

Recently, the most commonly recommended drugs for NeP in veterinary patients have been gabapentin, pregabalin and tricyclic antidepressants. The most frequently prescribed therapies for CM/SM are mainly gabapentin and NSAIDs 17, but there is little evidence in the veterinary literature about the efficacy and safety of the analgesic drugs used for NeP. Dosages may be adjusted, and other treatment options should be tried, if there is a lack of response, and it can be advisable to refer a patient to a pain specialist if the pain is not well controlled. Non-pharmacological options (e.g., electro-acupuncture) may be also considered alongside drug therapy.

Anticonvulsants

Gabapentin was developed as an anticonvulsant drug, but is being used to treat NeP in both humans and dogs. Gabapentin prevents the release of the excitatory neurotransmitter glutamate by blocking the voltage-dependent calcium channels. A beneficial effect in dogs with NeP has only been demonstrated in one study 23; the addition of gabapentin 10 mg/kg to carprofen improved the QoL (as assessed by VAS) compared to baseline and after carprofen introduction. A recommended dose of 10-20 mg/kg of gabapentin PO every 8 hours has been suggested, although as yet the range of plasma concentrations associated with clinical effects is uncertain. The most frequent adverse effects in humans include dizziness, somnolence, peripheral edema, weight gain, asthenia, headache and dry mouth.

Pregabalin has the same mechanism of action as gabapentin, and is preferred in humans because of its faster onset of analgesic effect; it is also more potent than gabapentin and produces fewer adverse effects. One pharmacokinetic study in dogs demonstrated that a dose of 2-4 mg/kg administered orally twice a day should provide adequate plasma concentration to reach the therapeutic range extrapolated from human studies 24. CKCS dogs with CM/SM have been treated successfully with pregabalin 25, and studies are currently in progress in an attempt to evaluate its analgesic effect for CM/SM. Topiramate is another antiepileptic drug which has various modes of action, including inhibition of carbonic anhydrase; this may lead to a reduction in CSF production, which in turn could reduce the size of the syrinx and potentially provide pain relief. However, in one study no significant differences were observed between topira mate and baseline or placebo 23 and owners reported that dogs treated with topiramate had a decreased appetite, a side effect also noted in human patients.

Antidepressants

Amitriptyline is considered a first-line therapeutic for NeP in humans. It is a serotonin and norepinephrine reuptake inhibitor, an NMDA antagonist, and voltage-gated sodium channel blocker. Additionally, amitriptyline enhances the activity of adenosine and GABA_B receptors and has anti- inflammatory effects. Anecdotal evidence from a case series reveals that two dogs with suspected NeP improved after amitriptyline administration 26.

Non-steroidal anti-inflammatory drugs (NSAIDs)

Prostaglandins modulate multiple sites along the nociceptive pathway and enhance both transduction of nociceptive information and transmission by the activation of glutamate and substance P in the spinal cord (central sensitization). Currently, there is no clinical evidence that NSAIDs reduce NeP. A one-week treatment with carprofen did not improve QoL (as assessed by VAS compared to baseline) in dogs with CM/SM in a crossover clinical trial 23.

Tramadol

Tramadol is considered as a weak opioid agonist, but may also inhibit reuptake of serotonin and noradrenaline at the level of the spinal cord to aid in the management of neuropathic pain. For this reason, tramadol should not be combined with antidepressants, but empirical doses of 1-5 mg/kg PO every 6-8 hours for dogs have been recommended for NeP. Opioid-like side effects (sedation and nausea) can be expected.

NMDA antagonists

Similar to ketamine, amantadine has N-methyl D-aspartate antagonist properties. It may reverse central pain sensitization and decrease tolerance to analgesics such as opioids. Amantadine given at 3-5 mg/kg PO every 24 hours could be effective in dogs with NeP, but the dosing frequency may need to be increased to every 12 hours because of its short half-life 27. A single case report described its use to treat NeP in a dog 28, but no controlled studies evaluating the analgesic effect specifically for NeP are available.

The progressive signs, severity of disease and high costs associated with diagnosis and treatment combine to generate substantial emotional and economic impacts for owners of dogs with CM/SM. More studies evaluating possible analgesics for the treatment of NeP are necessary in veterinary patients, as most of the medications currently employed are used empirically, with doses extrapolated from other species. A better understanding of the pathogenesis of NeP in CM/SM will also be advantageous for the best treatment selection in future.