Veterinary Surgery 37:186192, 2008

A Method for Intervertebral Space Distraction Before Stabilization Combined with Complete Ventral Slot for Treatment of Disc-Associated Wobbler Syndrome in Dogs
MERAV H. SHAMIR,
DVM, Diplomate ECVN,

ORIT CHAI,

DVM,

and EMMANUEL LOEB,

DVM

ObjectiveTo evaluate the use of a modied K-wire spacer for maintaining intervertebral distraction after ventral decompression and during stabilization as a treatment for disc-associated wobbler syndrome in large breed dogs. Study DesignA retrospective study. AnimalsDogs (n 7) with disc-associated wobbler syndrome. MethodsMedical records (20032006) of dogs treated by a modied surgical method were evaluated. Data retrieved were signalment, onset and duration of clinical signs, neurologic abnormalities, diagnostic methods, surgical procedure, immediate, and long-term ( 1 year) postoperative clinical and radiographic outcome. ResultsMean duration of clinical signs was 4.8 months. Neurologic signs included ataxia (2), ambulatory tetraparesis (2), and non-ambulatory tetraparesis (3). Three dogs had disc protrusion in 2 sites, 2 dogs had the procedure in 1 location and stabilization of both affected sites. All dogs improved dramatically and remained for 13 years. One dog had recurrence of cervical discomfort 13 months later. ConclusionsDespite the limited number of dogs, overall initial successful outcome with only 1 dog having mild recurrence 13 months later supports further use and evaluation of this technique. Clinical RelevanceDistraction using a K-wire spacer after ventral decompression followed by stabilization should be considered in dogs with disc-associated wobbler syndrome to prevent collapse of the intervertebral space. r Copyright 2008 by The American College of Veterinary Surgeons

INTRODUCTION ISC PROTRUSION in the caudal cervical spine of large breed dogs is known to be the most common of all 5 forms of caudal cervical spondylomyelopathy (CCSM) also known as wobbler syndrome.1 CCSM occurs primarily in large-breed dogs with Doberman and Great Danes most commonly represented.24 Other forms of CCSM are hypertrophy of the ligamentum avum, congenital vertebral malformation, vertebral tipping, and hourglass compression caused by hypertrophy of the articular facets and ligamentum avum.5,6 These forms differ in pathogenesis and direct cause of cord compres-

sion and thus require different surgical approaches for management. Clinical signs are typical to cervical myelopathy of varying degrees; ataxia and paresis of the pelvic limbs with hypometria of the thoracic limbs predominate but minor neurologic abnormalities like ataxia of the pelvic limbs alone or complete non-ambulatory tetraparesis are also seen. These signs are usually chronic and progressive but can deteriorate suddenly. Various surgical techniques have been reported for treatment of disc-associated wobbler syndrome including complete ventral slot, partial slot followed by distraction and stabilization, and distraction stabilization alone using different devices and implants.6,7 Most studies

From the Koret School of Veterinary Medicine, Hebrew University of Jerusalem, Israel. Address reprint requests to Dr. Shamir, DVM, Diplomate ECVN, Koret School of Veterinary Medicine, The Hebrew University of Jerusalem, PO Box 12, Rehovot 76100 Israel. E-mail: shamir@agri.huji.ac.il. Submitted March 2007; Accepted November 2007 r Copyright 2008 by The American College of Veterinary Surgeons 0161-3499/08 doi:10.1111/j.1532-950X.2007.00360.x

186

SHAMIR, CHAI, AND LOEB

187

report 7090% success rate with recurrence ranging from 2028% because of either postoperative collapse of the operated disc space, implant failure, or collapse of the adjacent disc termed the domino effect.1,3,4,610 All previously reported techniques either decompressed the cord by removal of disc material from the canal exposing the patient to subsequent collapse of the involved disc space or attempted decompression by distraction and stabilization, which may result in inadequate spinal decompression. Our purpose was to report the use of modied U-shaped K-wire as a spreader to maintain the desired intervertebral disc space as part of a combined surgical technique for treatment of disc-associated wobller syndrome in large breed dogs. We describe the use of a distracting device as a part of a surgical technique that includes complete ventral slot with removal of degenerated disc material from the spinal canal, insertion of cancellous bone graft, followed by stabilization using screws and polymethylmethacrylate (PMMA), and report outcome (clinical and radiographic) over 13 years. MATERIAL AND METHODS Inclusion Criteria
We retrieved medical records (20032006) of large breed dogs diagnosed with disc-associated wobller syndrome and treated using a combination of complete ventral slot, K-wire spreader for distraction, and screws with PMMA for stabilization. Dogs selected for the study had a myelographic diagnosis, surgical treatment, and  1 year follow-up. Both ambulatory and non-ambulatory dogs were included.

involved vertebrae (C4C5) were harvested, xed in 10% buffered formaldehyde, and then decalcied for 72 hours. Tissues were embedded in parafn, and processed by routine methods for microscopic examination using a hematoxylin and eosin stain.

Surgical Technique
Dogs were positioned in dorsal recumbency with the caudal part of the neck supported to achieve the desired degree of cervical extension. The head was secured to the table and the thoracic limbs were directed caudally with sufcient tension to apply linear traction on the cervical spine. Using a ventral approach to the cervical spine with extended elevation of the longus colli muscles from the vertebral bodies, the affected intervertebral disc and the entire ventral aspect of adjacent vertebral bodies was exposed. A high-speed drill was used to create a longitudinal full-thickness slot at the site of the intervertebral disc to remove the dorsal part of the anulus brosus and the hypertrophied dorsal longitudinal ligament to expose the spinal cord. The size of the slot did not exceed one-third the width and length of the involved vertebrae. Two 2.7 or 3.5 mm screws were inserted monocortically in each vertebra on both sides of the operated disc. Screw size and hole depth were selected individually for each dog based on CT or survey radiograph measurements of the involved vertebrae. The holes were drilled close to the midline on 1 side and then directed obliquely to the other side through the vertebral body to achieve maximal hole length. An angle of 301 to the midline axis was attempted for drilling and special care was taken to ensure that hole depth did not exceed the perpendicular height of the vertebral body to avoid penetration of the vertebral canal or the transverse foramina where the vertebral arteries are located (Fig 1). Distraction of the disc space was achieved by linear traction through pulling on the head and body and the use of a modied Gelpi retractor inserted into the slot. To maintain the intervertebral disc space width at 35 mm during stabilization, we inserted a K-wire spacer 2 or 3 mm diameter depending on vertebral size. Using 2 wire twisters, the wire was bent into a U shape so that the length of the horizontal part inserted into the slot was enough to provide the desired intervertebral space. One vertical arm was longer so that the spacer could be incorporated into the PMMA (Fig 1). Gelfoam soaked in morphine (0.5 mg/kg body weight) was inserted ventral to the K-wire to cover the spinal cord and supply sustained release local analgesia (Fig 1C). Cancellous bone graft harvested from the proximal humerus was inserted into the slot (above and around the K-wire) to enhance fusion of adjacent vertebrae (Fig 1C). Free fat graft was placed over the bone graft to avoid heat damage to the graft and provide additional protection to the spinal cord during the exothermic phase of acrylic hardening. PMMA was placed over the entire ventral aspect of both vertebrae with care to ensure that all 4 screws and the long arm of the modied U-shaped K-wire were incorporated and covered by acrylic. Saline (0.9% NaCl) solution was used to cool the PMMA as it hardened (Fig 1B, C).

Diagnostic Imaging
Myelographic diagnosis of spinal cord compression because of disc protrusion was made using 3 lateral and 1 ventrodorsal projections. The lateral projections were performed in traction, exion, and neutral position of the neck and were used to determine presence of a dynamic component. In 3 dogs, myelography was followed by computed tomography (CT) of the involved vertebrae. Immediate postoperative survey radiographs were obtained in all dogs and postoperative myelography and CT were performed in 1 dog.

Follow-up
Neurologic and radiographic evaluation was performed at least once for all dogs. Owners were invited for an additional follow-up examination of their dog before study completion to provide current information on clinical status and radiographic changes of the cervical spine. Long-term follow-up was 13 years.

Necropsy
One dog euthanatized 10 days after the surgery because of the detection of lung neoplasia was necropsied. The surgically

188

INTERVERTEBRAL SPACE DISTRACTION IN DOGS

distraction achieved. Follow-up radiographs were obtained every 612 months.

Neurologic Assessment
Neurologic assessment was scored by 1 neurologist (M.S.) initially and at each subsequent examination. Postoperative neurologic evaluation was done at 24 hours after surgery and on day 10. Dogs that had a longer recovery were evaluated monthly until outcome was considered satisfactory. Five dogs were reexamined 13 years after surgery by the same neurologist (M.S.) to acquire long-term outcome information whereas for 2 dogs (1 and 1.5 years after surgery) clinical status was determined by telephone communication with the owners.

Neurologic Scoring
The scoring method for cervical myelopathy was: 0 normal performance (may have slight low head carriage or very mild ataxia of the pelvic limbs); 1 pain or discomfort on manipulation of head and neck, stiff head carriage, cervical muscles spasm (with mild ataxia of the pelvic limbs); 2 ataxia or mild paresis of the pelvic limbs with normal thoracic limbs function; 3 ataxia involving all limbs with hypometric thoracic gait and hypermetric paresis in the pelvic limbs; 4 weakly ambulatory tetraparesisthe dog was able to rise and make few steps before collapsing; 5 non-ambulatory tetraparesis (no ability to rise or walk); and 6 non-ambulatory tetraparesis (tetraparesis with minimal to no voluntary movement detected either in the pelvic or thoracic limbs). Each dog was scored on admission and on each subsequent postoperative examination.

RESULTS Seven large breed dogs (5 males, 2 females; median age, 7.79 years; range, 5.510 years; mean weight, 33.4 kg; range, 3050 kg) were identied. Breeds were Weimaraner (4), Doberman (2), and Dog De Bordeaux (1). Four dogs (1, 4, 5, and 6) had disc protrusion at 1 site only whereas 3 dogs (2, 3, and 7) had an additional degenerated disc identied on CT myelogram, with disc space narrowing and sclerotic changes of the involved vertebrae but with milder compression (Fig 3A). In 2 of these dogs both diseased disc spaces were distracted and stabilized and the more severely affected site had surgical decompression as described. The third dog (2) had surgical decompression combined with distraction and stabilization for the more severely affected disc only. All dogs had satisfactory postoperative recovery with mean follow-up of 1.79 years (range, 13 years). No recurrence or deterioration of neurologic status was reported for 6 dogs whereas 1 dog (2) that had surgery at C5C6 deteriorated 13 months later because of disc protrusion at C6C7. This was the disc that had apparent slight degeneration on preoperative radiographs. This

Fig 1. Preoperative (A) and postoperative (B) lateral radiographs and schematic drawing (C) of a technique using a modied K-wire spacer (Cblack line) to maintain intervertebral space after complete ventral slot until the polymethylmethacrylate (PMMA) hardens. Note the position of morphine soaked Gelfoam (Cdark gray rectangle) and cancellous bone graft (Cdotted area). The PMMA (Clight colored ellipse) cover the screws and the long arm of the K-wire spacer completely.

Radiographic Assessment
Postoperative lateral and ventrodorsal projections were taken to document nal screw position and the amount of

SHAMIR, CHAI, AND LOEB

189

dog improved gradually after 4 weeks of connement followed by 3 months of restricted activity. Two other dogs (3, 7) with narrowing and protrusion of 2 adjacent discs, responded well to double space stabilization with the most protruded disc decompressed. Outcome at 2.5 years was good in 1 dog (score improved from 3 to 2) and was excellent at 1 year for the other dog (score improved from 5 to 1). Five dogs had improvement within 48 hours of surgery whereas the other 2 (dogs 2 and 3) had a slower recovery. These dogs were admitted 18 and 6 months after onset of clinical signs and were scored 4 and 3 for neurologic dysfunction on admission. Six dogs had complete recovery with return to normal activity although some limitation of cervical range of motion was evident at follow-up neurologic evaluation in 2 dogs (1, 2). One dog (6) was admitted with severe lateral recumbent tetraparesis (score 6) and improved dramatically, and was standing unaided within 48 of surgery. This dog took longer to fully return to walking but gradually improved over the next 3 months. In the last followup telephone conversation with the owner 1 year later, the dog was still improving. Radiographic Outcome Beyond initial postoperative radiographs and at least 1 set taken within the rst year, 5 dogs had an additional 2 5 radiographic examinations during follow-up. Increased opacity of the slot, indicating different levels of fusion for operated vertebrae was detected at 912 months in all dogs. No collapse of the gap between vertebrae was detected indicating that the stabilization technique used was sufcient to maintain the required gap until the vertebral fusion (Fig 3B, C). Slowly progressive degenerative changes represented as narrowing of the disc spaces and sclerotic changes in the vertebral endplates were observed in the adjacent disc spaces of all dogs examined. If surgery was performed initially at the C5C6 disc space, the next disc that had degenerative changes was always C6C7 whereas when C6C7 disc had initial surgery, the C5C6 space was always next to have postoperative degenerative change (Fig 3C). Changes were progressive and included narrowing of the disc space, sclerotic changes of the vertebral endplates, and osteophyte formation and were detected at 1-year follow-up and clearly evident 1.53 years after initial surgery (Fig 3C). DISCUSSION The technique we report, a combination of complete ventral slot that includes removal of the degenerated disc from the spinal canal, distraction using a K-wire spacer

until the PMMA hardens, and stabilization with screws and PMMA, is a modication of an earlier method for treating disc associated wobbler syndrome in large breed dogs. Although our case experience was small, we believe that this combined procedure using a modied U-shaped K-wire to achieve and maintain the desired space between the involved vertebrae was effective. The combination of complete ventral slot with distraction and stabilization permits careful removal of disc material from the canal enabling the surgeon to view the spinal cord and subsequent distraction and stabilization of adjacent vertebra to prevent collapse of the space or joint instability. Insertion of a modied U-shaped K-wire into the slot helps achieve and maintain the desired amount of parallel distraction of the disc space until the PMMA hardens. The long arm of the K-wire implant is then incorporated in the PMMA, eliminating the concern of future wire migration. The K-wire spreader is prepared to t the specic slot size in each patient after distraction using a modied Gelpi retractor inserted in the slot. Other methods of maintaining the intervertebral space have been reported such as stainless-steel or titanium cages, or washer and screws11; however, these devices result in continuous compressive loads on the vertebral end plates. The bone eventually remodels around the metal implant and engulfs the device, which can result in collapse of the intervertebral space and recurring bulge and may again compress the spinal cord.11,12 Our method avoids this complication because the spacer is fully loaded only during the time required for the PMMA to harden after which most of the load is taken by the screws and PMMA. A modied Gelpi retractor where the sharp ends were cut on both sides is commonly used to temporarily distract the intervertebral space. Correct placement of this modied Gelpi is not always simple and requires exposure of the ventral aspects of additional vertebrae and fenestration of the adjacent discs. Additional soft-tissue damage to muscles and ligament weakens the neighboring intervertebral joints and may result in instability and recurrence of signs.11 The modied spacer is simple to construct from K-wire and is inexpensive and readily accessible. Furthermore the spacer occupies only a small amount of the slot space enabling the surgeon to t the device into the slot while the modied Gelpi retractor is in position. Once the spacer is positioned, the Gelpi retractor can be gently removed leaving ample space for insertion of large quantities of cancellous bone graft with good contact to the exposed vertebral bone to ensure later vertebral fusion (Figs 1 and 3). During the exothermic phase of PMMA hardening, heat is transmitted through the metal implants (screws, K-wire) and may damage the vertebral bone surrounding the screws and the cancellous graft. Tissue from the dog

190

INTERVERTEBRAL SPACE DISTRACTION IN DOGS

Fig 2. Histologic examination of cancellous bone graft from a dog that died 10 days after the surgery. The allograft is composed of a fair amount of brin and erythrocytes (A and B). Adjacent to the vertebral bone (A) some intact, spindle shaped cells (A and B arrows) were observed along woven bone trabeculae. Scalebar 20 lm.

that died 10 days after surgery had mild coagulation necrosis marked with a sharp demarcation zone between necrotic and vital bone tissue in the area surrounding the screws on both C5 and C6 vertebrae, and the bone allograft was composed of fair amount of brin and erythrocytes. In areas adjacent to the vertebral bone some intact, spindle-shaped cells, considered activated osteoblasts were observed along woven bone trabeculae (Fig 2B). These ndings suggest that the bone damage caused by heat produced from the PMMA and transmitted through the K-wire and screws was minimal and based on follow-up radiographs for other dogs did not prevent vertebral fusion (Fig 3C). Care must be taken to ensure that the size and shape of the PMMA are suitable to allow complete coverage of all screw heads and the long arm of the K-wire spacer. A larger piece of cement or a protruding sharp end of the spacer should be avoided because they might interfere with normal function of the esophagus and trachea.

Fig 3. Preoperative (A), postoperative (B), and 15-month follow-up (C) lateral radiographs of dog 1 (Table 1). Note the amount of distraction achieved at C5C6 intervertebral space using the K-wire spacer (B) and the maintenance of implants position at 15 months (C). Degenerative changes of the discs spaces on both sides of the operated site are clearly evident (C) although narrowing of the C6C7 disc space was present preoperatively (A).

Two dogs, one non-ambulatory (dog 7) and another weak tetraparetic (dog 1) were able to stand and walk steadily the morning after the surgery despite a relatively long time between onset of clinical signs and surgical intervention (3 and 1 month, respectively). This may reect meticulous removal of foreign material from the spinal

SHAMIR, CHAI, AND LOEB

191

Table 1. Summary Data for 7 Dogs with Disc-Associated Wobbler Syndrome Treated by Ventral Slot Decompression, Distraction, and Stabilization Dog 1 2 3 4 5 6 7 Age (years), Sex, Breed 8 M Doberman 10 M Weimaraner 5.5 F Weimaraner 7 M Doberman 9 M Weimaraner 7M Weimaraner 8 F Dog de Bordeaux Duration (months) 1 18 6 0.5 3 1 1.5 Admission Neurologic Score 4 4 3 3 5 6 5 Lesion C5C6 C5C6 C6C7 C5C7 C6C7 C6C7 C5C6 C6C7 C5C6 C6C7 Follow-up (years) 3 2.5 3 1.5 1 1 1 Follow-up Neurologic Score 0 2 2 1 0 3 1 Recurrence none C6C7 none none none none none

M, male; F, female.

canal followed by efcacious distraction of the disc space using the K-wire spacer before stabilization. This combined surgical procedure may have potential long-term advantages. Any stabilization device applied after distraction of vertebrae in the caudal cervical area may fail because of the load concentrated on relatively small areas of contact between the bone and implant. This load may result in local pressure necrosis of the bone surrounding the stabilizing implant (e.g. screws/washer/ cement plugs) on both sides of the gap and result in collapse of the disc space.13 If this occurs when complete ventral slot was not performed, rebulging of any remaining degenerated anulus brosus and the hypertrophied dorsal longitudinal ligament will result in spinal cord compression. This may account for the relatively high prevalence (10%) of early deterioration in neurologic status reported in several studies14,15 and some of the more delayed relapses of clinical signs that exceeded 30% in other studies.12,13,16, We performed complete ventral slot before distraction to eliminate this cause of recurrent spinal cord compression at the operated site. All of our dogs improved within the rst week after surgery and continued to improve during the rst few months. Six dogs had no recurrence of neurologic dysfunction during follow-up of 13 years. Dog 2, admitted for surgery 18 months after initial clinical signs, improved dramatically and functioned normally for 13 months before recurrence of neurologic signs. CT myelography at that time revealed disc protrusion at the intervertebral disc space adjacent the operated space. This outcome is referred to as the domino effect. This disc space had evidence of degeneration on preoperative radiographs but we decided not to distract or stabilize this space. The domino effect is believed to be caused by load transmission from the stabilized intervertebral disc to the interspace immediately cranial or caudal to it. Load transmission exacerbates subclinical instability in an adjacent disc space enhancing the degenerative processes that lead to disc protrusion and dorsal longitudinal

ligament hypertrophy.17 Although only 1 dog developed a domino effect, we do not believe our technique has any advantage over previously reported techniques in preventing domino effect. It would appear that distracting and stabilizing both the clinical and subclinical degenerated disc space may be benecial in preventing recurrence of spinal compression from the domino effect. In the other 2 dogs (3, 7) with degenerative changes in adjacent discs spaces, ventral decompression was performed at the most affected site but distraction and stabilization was performed at both clinical and subclinical disc spaces. Neither of these dogs had recurrence of clinical signs during follow-up of 1 and 3 years, respectively. Mild degenerative changes of C4C5 were observed 2 years later in dog 3 that had both C5C6 and C6C7 stabilized (Table 1); however, no clinical deterioration occurred over the next 3 years. Extensive degenerative changes in neighboring disc spaces on both sides of the operated disc indicate excessive instability in interspaces that are now carrying the load transmitted from the operated intervertebral space.17 In some cases these degenerative changes may result in ankylosis of the disc space or spaces without causing the dog any apparent discomfort. Although we report a small number of cases, our technique was benecial in both ambulatory and nonambulatory dogs with no recurrence in 6 of 7 dogs with a mean follow-up of 1.9 years. Improved understanding of the forces and loads affecting the caudal cervical region in large breed dogs will facilitate development of more suitable methods to minimize development of further degenerative changes that leads to recurrence of neurologic signs. REFERENCES
1. Queen JP, Coughlan AR, May C, et al: Management of discassociated wobbler syndrome with a partial slot fenestration and position screw technique. J Small Anim Pract 39:131136, 1998

192

INTERVERTEBRAL SPACE DISTRACTION IN DOGS

methacrylate plug in dogs with caudal cervical spondylomyelopathy. J Am Vet Med Assoc 208:6168, 1996 Wilson ER, Aron DN, Robert RE: Observation of a secondary compressive lesion after treatment of caudal cervical spondylomyelopathy in a dog. J Am Vet Med Assoc 205:12971299, 1994 Rusbridge C, Wheeler SJ, Torrington AM, et al: Comparison of two surgical techniques for the management of cervical spondylomyelopathy in Dobermans. J Small Anim Pract 39:425431, 1998 McKee WM, Lavelle RB, Richardson JL, et al: Vertebral distraction-fusion for cervical spondylopathy using a screw and double washer technique. J Small Anim Pract 31: 2227, 1990 Chambers JN, Oliver JE, Bjorling DE: Update on ventral decompression for caudal cervical disk herniation in Doberman pinchers. J Am Vet Med Assoc 22:775778, 1986 Ellison GW, Seim HB, Clemmons RM: Distraction cervical spinal fusion for management of caudal cervical spondylomyelopathy in large-breed dogs. J Am Vet Med Assoc 193:447453, 1988 Bruecker KA, Seim HB, Blass CE: Caudal cervical spondylomyelopathy: decompression by linear traction and stabilization with Steinmann pins and poly-methylmethacrylate. J Am Anim Hosp Assoc 25:677683, 1989 Hilibrand A, Carlson G, Palumbo M, et al: Radiculopathy and myelopathy at segments adjacent to the site of the previous anterior cervical arthrodesis. J Bone Joint Surg (Am) 81:519528, 1999

2. Mason TA: Cervical vertebral instability (wobbler syndrome) in the dog. Vet Rec 104:142145, 1979 3. Jeffery ND, McKee WM: Surgery for disc-associated wobbler syndrome in the dogan examination of the controversy. J Small Anim Pract 42:574581, 2001 4. De Risio L, Munana K, Murray M, et al: Dorsal laminectomy for caudal cervical spondylomyelopathy: postoperative recovery and long-term follow-up in 20 dogs. Vet Surg 31:418427, 2002 5. Bruecker KA, Seim HB: Caudal cervical spondylomyelopathy, in Slatter D (ed): Textbook of Small Animal Surgery (ed 2). Philadelphia, PA, Saunders, 1993, pp 10651079 6. Sharp NJH, Wheeler SJ: Cervical spondylomyelopathy, in Sharp NJH, Wheeler SJ (eds): Small Animal Spinal Disorders Diagnosis and Surgery (ed 2). Edinburgh, UK, Elsevier Mosby, 2005, pp 211232 7. Voss ER, Steffen F, Montavon PM: Use of the compact unilock system for ventral stabilization procedures of the cervical spine. A retrospective study. Vet Comp Orthop Traumatol 19:2129, 2006 8. Dewey CW: Myelopathies: disorders of the spinal cord, in Dewey CW (ed): Canine and Feline Neurology (ed 1). Ames, IA, Iowa State Press, 2004, pp 277336 9. Bruecker KA, Seim HB, Withrow SJ: Clinical evaluation of three surgical methods for the treatment of caudal cervical spondylomyelopathy of dogs. Vet Surg 18:197 203, 1989 10. Dixon BC, Tomlinson JL, Kraus KH: Modied distractionstabilization technique using an interbody Polymethyl

11.

12.

13.

14.

15.

16.

17.