Cervical Stenotic Myelopathy

Robert J. MacKay

The term wobbler (also wobbler syndrome) describes horses with ataxia caused by any of a group of developmental anomalies and degenerative conditions involving the cervical vertebrae.¹ Several other descriptions are used in wobbler terminology.

Two overlapping CSM syndromes are recognized, differing greatly in developmental histories. Type I is characterized by focal or multifocal stenosis of the vertebral canal, principally between C1 and C6, that is worse on flexion (dynamic stenosis). Affected horses are typically younger than 2 years of age. In type II, bony and soft tissue impingement into the vertebral canal results from remodeling of the articular process joints of the caudal cervical vertebrae (usually C5 to T1) and stenosis of the cranial vertebral orifice. This impingement is present in all positions of the neck (static stenosis) but may also have a dynamic component in that it worsens during hyperextension. Horses with static stenosis are usually young to middle-aged at the onset of wobbler signs.

Both syndromes probably involve some combination of aberrations in cartilage and bone development and external trauma that may precede clinical ataxia by many months (perhaps even in utero in the case of type I CSM¹).

■ Pathogenesis Type 1 CSM is another manifestation of the group of disorders of endochondral ossification (i.e., osteochondroses) that occur in young, growing horses and include physitis, contracted tendons, osteochondritis dissecans, subchondral bone cysts, and malformation of the cuboidal bones of the carpus and tarsus.³ Study of wobbler matings has revealed a propensity for rapid growth and superior performance among offspring but no clear evidence that CSM was genetically predetermined.⁴ Of interest is that other developmental orthopedic diseases were overrepresented.

As a result of presumed genetic, physical, and dietary factors, osteochondrosis of the articular processes and metaphyseal growth plates occur early in affected foals. Genetic predisposi­tion, rapid growth, maternal nutrition, calcium/phosphorus imbalance, low dietary levels of copper, and high dietary levels of zinc have all been implicated as developmental orthopedic diseases in foals, but their roles as risk factors for CSM are not clear.^6-9 Osteochondrosis of the vertebral physes of rapidly growing foals and weanlings results in multiple vertebral abnormalities, including flaring and enlargement of the caudal metaphyses and epiphyses of the vertebral bodies (equivalent to physitis of the limbs) and disparity of longitudinal growth of the vertebral bodies and dorsal laminae. These deformities cause focal stenosis of the vertebral canal and susceptibility to dynamic compression, respectively. In severe forms, the latter abnormality forces pairs of vertebrae into a permanently flexed, malarticulated, kyphotic position in which severe vertebral canal stenosis (both static and dynamic) is likely to occur. Flaring of the caudal metaphyseal-epiphyseal unit contributes to the narrow caudal orifice of the canal within an affected vertebra that is common in horses with CSM.

Osteochondrosis of the articular process joints may result in joint instability and later osteoarthritis and remodeling. Associ­ated with degenerative joint disease of the articular process joints are gross enlargements due to osteophyte development.

■ Necropsy Findings Horses with CSM have one or more focal compression-type lesions in the cervical spinal cord.¹⁰ The gross appearance of such lesions is of flattening and palpable softening of the cord. They are characterized histologically by swollen and disrupted axons and phagocytosis of myelin in white matter. Mild hemorrhages are sometimes observed in acute lesions. In chronically affected animals, the white matter demonstrates continued degeneration of neuronal fibers, proliferation of capillaries with prominent fibrous coats, and astrofibrosis. More severely affected horses also have neuronal necrosis, loss of cell bodies, and sometimes astrofibrosis in the gray matter. Massive focal lesions also occur with cavitating necrosis of white matter, particularly in lateral funiculi and occasionally in gray matter. Markedly asymmetric lesions are found in horses with prominently asymmetric clinical signs. Cranial to the focal lesions, neuronal fibers (axons and myelin) in ascending (proprioceptive) tracts exhibit secondary degenera­tion for variable distances, depending on the extent and age of the focal lesion.

According to one report, the most common histologic lesions in the articular processes of horses with CSM were osteochon­drosis, osseous cyst-like structures, fibrous tissue replacement of trabecular bone, retained cartilage matrix spicules, bone cysts, and osteosclerosis.¹¹

■ Clinical Signs Most animals with type I CSM are 6 to 18 months old, whereas those with type II CSM are typically 18 months to 5 years; however, there are many exceptions to these ranges, and both conditions can occur in a single animal. In four large breeding farms, the onset of type I CSM occurred in horses aged 202 to 986 days (average, 434 days), whereas type II CSM cases began at ages 1085 to 1299 days (average, 1188 days).¹² A study of 811 horses in the United States indicated that horses of the Thoroughbred, warmblood, and Tennessee Walking Horse breeds were significantly more likely to have CSM than were Quarter Horses; Standardbreds and Arabians were significantly less likely to have CSM.¹³ Large warmblood breeds are notably affected by type II CSM.¹⁴ Males are over­represented.^14,15 Affected horses are often well grown, and some have a history or signs of other developmental orthopedic diseases such as osteochondritis dissecans or physitis.

Clinical signs of ataxia and paresis typically begin insidiously but may manifest suddenly. In the latter case, there is often a history of some minor fall or training incident that confuses the diagnosis (the fall could have been a precipitating cause or a result of the disease). After signs have been recognized, they typically progress for a period of months and then may stabilize or even regress.

The principal neurologic signs are ataxia, weakness, and spasticity of the limbs. These signs reflect involvement of axons of upper motor neurons and general proprioceptive tracts in the white matter of the spinal cord underlying the site or sites of compression. Because of the more superficial location of general proprioceptive tracts from the pelvic limbs, the pelvic limbs are generally more affected than the thoracic limbs.¹ Dynamic stenosis is usually associated with symmetric signs, whereas static stenosis may cause asymmetric signs. Severe arthrosis of the joints between the C6 and C7 levels or between the C7 and T1 levels occasionally causes signs more severe in the thoracic than in the pelvic limbs. The first sign noticed may be stumbling or clumsiness when the horse is moving into or out of a box stall. During exercise gallops, the rider may comment on a feeling of looseness or “losing the back end” around turns. Reluctance to change leads, cross-cantering or cross-galloping, or even falls after stumbling are other early signs in horses being ridden. With progression, affected horses may exhibit gait at the walk that is obviously stiff and unco­ordinated; the prominent truncal sway; or “wobble,” which explains the common name for CSM in horses. Because of pelvic limb extensor weakness, some wobblers catch their patellas or have persistent upward patella fixation.

In addition to the signs already mentioned, careful neurologic examination usually discloses some or all of the following:

• Abnormal pelvic movements during walking, best observed from behind, including excessive swiveling of an imaginary line on the median plane from between the tuber sacrale to the tailhead, excessive rotation of the pelvis around its long axis (visible as exaggerated amplitude of the rise and fall of the tuber coxae), and side-to-side truncal sway.

• Decreased resistance to lateral pressure exerted by pulling of the tail while the horse is walking.

• Apparent stiffness of the limbs during the swing phase of the stride in the thoracic or pelvic limbs. This may be associated with scuffing of the toes. When the horse's head is raised during walking, especially down a gentle slope, the stiff hypometric gait in the thoracic limbs may be exaggerated and is sometimes described colloquially as a “marching,” “floating,” or “tin-soldier” gait.

• Circumduction of each outside pelvic limb when the horse is walked in small circles.

• Interference between the thoracic limbs and pivoting around the pelvic limbs when the horse is moved in tight circles.

• Absence of the thoracolaryngeal (“slap”) reflex.

In horses with markedly asymmetric compression (usually horses with asymmetric caudal intervertebral arthritis), there also may be marked asymmetry of stride length between the pelvic limbs.

■ Diagnosis Radiography of the cervical spine is the most important diagnostic tool. CSM can usually be diagnosed presumptively on plain laterolateral radiographs of the cervical spine from the finding of focal stenosis of the vertebral canal. The diagnosis is buttressed by additional demonstration of malformation, osteochondrosis dissecans, or arthrosis of the articular process joints; metaphyseal and epiphyseal flaring of the caudal aspect of vertebral bodies; relative elongation caudally of the dorsal laminae; or angulation between pairs of vertebrae. Oblique projections (usually lateral-50 degrees-ventral to laterodorsal) enable more accurate description of remodeling and other changes in the articular process joints and should be included as “standard of practice” in studies of horses with arthroses of the articular process joints.¹⁶ Standardized measure­ments of articular process joints in oblique projections have been published, but poor repeatability and inaccuracy of these measurements have so far limited their usefulness.¹⁷

Intravertebral sagittal ratios for vertebrae behind C2 are calculated as the ratio of minimal sagittal diameter of the vertebral canal to maximal sagittal diameter of the cranial half of the vertebral body].¹⁸ Both measurements are made per­pendicular to the long axis of the vertebral canal. The denomina­tor normalizes for variations in horse size and distance between object and image. Values for sagittal ratio of less than 0.52 from C4 to C6 and less than 0.54 at C7 are predictive of spinal cord compression with 89% sensitivity and specificity in the diagnosis of spinal cord compression, according to the original description.¹⁸ In practice, cutoff values of 0.50 for C3-C6 and 0.52 for C7 are typically used.¹ Even horses with static stenosis, which may produce predominantly transverse compression from large joints, usually have at least one abnormal intraver- tebral sagittal ratio.¹⁸

Intervertebral sagittal ratio measurements are made between pairs of vertebrae in order to capture the effect of fixed flexion or extension angles on the diameter of the vertebral canal.¹⁹ For a pair of vertebrae fixed in a kyphotic angle, the primary measurement (numerator) is the shortest distance between the caudal aspect of the dorsal lamina and the craniodorsal aspect of the vertebral body of the caudal-adjacent vertebra. For a fixed lordotic intervertebral joint, the corresponding measure­ment is the shortest distance between the caudodorsal aspect of the vertebral body and the most cranial point on the dorsal lamina of the caudal-adjacent vertebra. This measurement is then divided by the maximal sagittal diameter of the cranial half of the vertebral body of the cranial-adjacent vertebra to calculate the intervertebral sagittal ratio.¹⁹ Normal values for means and lower bounds of the 95% confidence intervals for intravertebral and intervertebral sagittal ratio from C3 to C7 have been published.^1,19 Any value below these lower bounds is highly predictive of spinal cord compression.

A semiquantitative scoring system has been developed for use in foals 3 to 7 months of age to predict the risk of developing CSM.²⁰ Intravertebral and intervertebral sagittal diameters are recorded as described earlier, but vertebral body length (rather than sagittal diameter) is used as denominator to render these diameters as ratios.²⁰ Up to 10 points are assigned for decreased sagittal ratios and up to 5 points for each of the following signs of cervical vertebral malformation: dorsal flaring of the caudal vertebral body, caudal extension of the dorsal lamina, angulation between adjacent vertebral bodies, delayed ossifica­tion, and enlargement of articular processes. A score of 12 or higher (maximum, 35) is predictive of the development of CSM. This study revealed a strong tendency for the diameter of the vertebral canal to be greater in female foals than it was in males.²⁰ A multicomponent and detailed system of measure­ments in plain and myelographic studies devised by Japanese workers^21,22 accurately identifies foals and yearlings with CSM, but it has not been widely adopted in other countries.

When plain radiographs are abnormal, contrast myelography may be used to support the diagnosis, to establish the number and locations of sites involved and the severity of lesions in each, and to define the soft tissue contribution to spinal cord compression. These factors are especially important if surgical stabilization is contemplated because myelography may reveal a site of disease different from that suspected from plain films. With the horse under general anesthesia, 30 to 60 mL of iohexol (Omnipaque) or iopamidol (Isovue) radiocontrast solution are injected slowly into the atlantooccipital subarachnoid cistern after a similar volume of CSF has been removed. After injection has been completed, the head is raised for 5 minutes to facilitate caudal flow of dye, and then the head is lowered and radiographs are taken.

Laterolateral views of the cervical spine are taken in neutral, flexed, and extended positions. Spinal cord impingement due to dynamic compression is clearly defined in lateral views if the dorsal and ventral dye columns are completely extinguished at diametrically opposed sites in flexed views. For cases of incomplete compression of the dorsal column, several different diagnostic criteria have been used, none of which is completely satisfactory.^23,24 The traditional standard of complete compres­sion of the ventral dye column and 50% or more compression of the dorsal column at diametrically opposed sites is sensitive but has poor specificity (i.e., it can yield many false-positive results). Other likely, more accurate alternatives are (1) com­parison with reference ranges for minimal dural diameters and minimal flexed dural diameters for horses either above or below 320 kg and (2) 30% or greater reduction in dural diameter in neutral views at the site of interest (usually intervertebral) in comparison with the midbody dural diameter in the caudal- adjacent vertebra. Sites affected in order of frequency are the C3-C4, C4-C5, and C5-C6 levels. Multiple sites may be affected in a single horse. The typical myelographic feature of static stenosis is focal spinal cord compression throughout the full range of vertebral motion. Any dynamic component at these caudal sites is relieved by flexion of the spine. The C6-C7 articulation is involved most frequently, and C5-C6 and C7-T1, to a lesser extent. At these sites, absolute narrowing of the minimal dural diameter or 30% reduction in comparison with the cranial-adjacent midbody measurement are the best diagnostic criteria.¹ In cases in which the plane of compression is transverse rather than dorsoventral, the dural diameter is usually abnormal; however, less objective criteria may have to be considered when measurements are normal. These include widening of the spinal cord diameter and fading of the dye columns over the area of lateral compression. Unfortunately, the diagnostic accuracy of plain radiographic or myelographic diagnoses of CSM has been unimpressive in comparison with that of postmortem spinal cord histopathologic study.¹⁵

Advanced imaging with CT²⁵ or MRI²⁶ have been described, but in mature horses, gantry dimensions usually limit their use to vertebrae in front of C4. The current rapid pace in development of MRI and CT equipment capable of evaluating the cervical spine of mature horses may in the near future substantially enhance the evaluation of CSM.

A technique has been described for endoscopic evaluation of the epidural and subarachnoid spaces of the cervical spinal cord, and this procedure was applied to a horse with CSM.^27,28 Although the technique may have application in selected cases, it has no place in the routine diagnosis of CSM.

Needle electromyography is most useful for investigation of muscles deprived of lower motor neuron activity, but it has little application to the diagnosis of CSM. The most promising electrodiagnostic testing application in this setting is transcranial magnetic stimulation., Descending motor tracts are stimulated by current pulsation from a coil placed on the scalp. The onset latency of electromyographic responses in limb muscles is related to the integrity of spinal motor pathways. Latencies for horses with CSM were significantly longer than those for normal horses.³¹

CSF analysis generally yields normal results. At most, xanthochromia and elevated protein levels (70 to 130 mg/dL) are observed.

■ Treatment

MEDICAL. Corticosteroids and NSAIDs may provide transient improvement in the signs of CSM but do not treat the underlying problem.

Large degenerative intervertebral joints can be seen ultra­sonographically, and corticosteroid (e.g., triamcinolone or methylprednisolone acetate) can be injected intraarticularly under ultrasound guidance, as has been described.³² The theo­retical indications for this procedure include relief from pain of arthritis and reduction in the soft tissue components of arthritic joints so as to provide at least temporary reversal of their impingement on the spinal cord and spinal nerves in the case of CSM and thoracic limb lameness, respectively. Although descriptive support for all of these possibilities was provided in an observational series,³³ no fully objective evidence has been provided for reversal of the signs of CSM, and it is inconceivable that permanent relief could be provided.

Some mildly affected horses can stabilize and even improve both radiographically and clinically if confined and placed on restricted but nutritionally complete feed.¹ There are reports of complete recoveries with this conservative approach and, in one report, 30% of conservatively treated mildly affected Thoroughbreds went on to race.³⁴ For affected horses younger than 1 year (and preferably younger than 6 months), a restrictive “paced-growth” diet can be tried in order to retard growth rate and prevent CSM in clinically normal foals with radiographic evidence of stenosis of the vertebral canal or to treat foals with mild clinical signs of CSM.³⁵ This diet reduces energy intake to 65% to 75% of National Research Council recommendations and increases levels of trace minerals. The essential elements are as follows: (1) immediate weaning if the foal is still suckling and (2) feeding free-choice grass hay (e.g., coastal Bermuda hay) plus 3.5 lb daily of a ration balancer that is high in micronutrients and low in calories (e.g., Buckeye's Gro 'N Win). At 1 year of age (based on birth date), the supplement is reduced over 1 week to 1 lb daily. Foals should be confined in a stall except for limited round-pen or equivalent turnout time, certainly without access to pasture. Good results are claimed for slightly affected foals, although breaking, training, and racing are set back several months and the foals look poor as yearlings.³⁵

SURGICAL. Ventral interbody vertebral fusion is the principal procedure. In this surgery, pairs of vertebrae are fused in exten­sion by implantation of a cylindrical stainless steel or titanium basket (Cloward Bagby basket) filled with autogenous cortical bone fragments.^36,37 Osseous union of the vertebral bodies and stabilization of the articular process joints occurs after this surgery, although nonunion is an occasional postoperative problem. A modification involves use of a threaded titanium implant that screws into a prepared circular, threaded kerf around an isthmus of preserved bone. By leaving the isthmus and its blood supply intact, it is expected that complete bony fusion occurs more quickly. A plate may be secured over the implant, but this does not prevent retrograde migration of the implant.

The surgery provides immediate relief from dynamic stenosis and may improve static stenosis over months by causing atrophy of arthritic intervertebral joints. In the latter case, intraarticular administration of corticosteroid may hasten atrophy of the enlarged joint. The surgery results in improvement in neurologic status in 44% to 90% of horses with dynamic compression and 12% to 62% of treated horses return to athletic function.^36-39 Overall, the success rate offered by most surgeons is approxi­mately 60%, with normal myelographic columns expected by 2 months after surgery and complete osseous union by 1 year postoperatively. On average an improvement of 1 to 2 grades is expected.³⁸ Results of surgery for dynamic compression were significantly better than those for static compression in one published series.⁴⁰ Although there is a general impression that increasing time since onset of signs, higher neurologic grade, and multiple surgical sites are negative indicators for response to stabilization surgery, published results are not consistent with regard to any of these issues.³⁶ Even three-site surgeries can be successful, although neck mobility is diminished.⁴¹ The use of a locking compression plate has been described with initial promising results.³⁶ If the results of locking-plate surgery continue to be positive, such procedures may become more widely available, obviating the need for the specialized equip­ment needed for implant surgery.³⁶

Dorsal subtotal decompressive laminectomy⁴² provides imme­diate relief for static compression but technically is extremely difficult and fraught with severe postoperative complications.