Effectiveness of Cervical Hemilaminectomy in Canine Hansen Type I and Type II Disc Disease: A Retrospective Study
Medical records of 41 dogs, including 15 small breed dogs (<15 kg) and 26 large breed dogs (>15 kg), with cervical intervertebral disc disease (IVDD) that underwent a hemilaminectomy were reviewed. Dogs were diagnosed using myelography, computed tomography/myelography, or MRI, and dogs were classified as having either Hansen Type I disc extrusion or Hansen Type II disc protrusion located ventrally, ventrolaterally, or laterally within the cervical spinal canal. The most common clinical presentation was ambulatory tetraparesis and/or lameness (44%). The most affected sites for cervical IVDD were between the sixth and seventh cervical vertebrae (C6–C7; 78% of Hansen Type II discs) and C2–C3 (86% of Hansen Type I discs). Treatment was effective in 88% of dogs. Five large breed dogs (12%) did not improve. In dogs with a Hansen Type I disc extrusion, clinical signs improved in 96% of the cases. In dogs with a Hansen Type II disc protrusion, an excellent and good outcome was seen in 47% and 32% of cases, respectively. Outcome was significantly better for small breed dogs and dogs with Hansen Type I disc disease compared with large breed dogs and dogs with Hansen Type II disc disease.
Introduction
Cervical intervertebral disc disease (IVDD) accounts for 15% of all intervertebral disc problems in dogs.1 Two forms of disc disease that may require surgical management have been identified: Hansen Type I (extrusion of the nucleus pulposus) and Hansen Type II (continued degeneration and protrusion of the outer fibrous layers of the annulus).2–4 Several spinal decompression techniques have been described and evaluated for the cervical spine, including ventral slot decompression with and without subsequent stabilization and dorsal decompressive laminectomy. Although these procedures are effective, they can be associated with various complications, including hemorrhage from the vertebral plexus, residual material within an intervertebral foramen, luxation and instability at the surgical site for the ventral slot, the extent of muscle dissection required during the approach, and difficulties addressing ventral and ventrolateral disc material during dorsal decompression.5 Potential benefits of cervical hemilaminectomy include good visualization of the lateral spinal cord, nerve roots, and spinal canal, which facilitates removal of disc material in these areas because it does not remove large portions of either the intervertebral disc or vertebral body. Further, hemilaminectomy may avoid destabilization of the operated segment.
To date, the use of hemilaminectomy in the cervical spine has been described mainly for lateralized disc herniations, laterally located masses such as neoplasms or synovial cysts, or for lesions at the seventh cervical vertebra (C7) to the first thoracic vertebra (T1).6–10 There is only limited information in the literature regarding access to disc material located ventrolaterally and ventrally to the spinal cord by hemilaminectomy.10,11 The current study provides data regarding one institution's hemilaminectomy results in a population of dogs suffering from various forms, and different compression sites, of cervical IVDD (Hansen Type I and Hansen Type II) and disc-associated wobbler syndrome (DAWS).
Materials and Methods
Inclusion Criteria
The medical records of 41 dogs that underwent hemilaminectomy for the management of IVDD in the cervical spine at the Clinic for Small Animal Surgery/Neurology inXZurich, Switzerland between 2003 and 2007 were reviewed. Retrieved information included breed, age, sex, presurgical neurologic status, type of imaging modality (myelography, computed tomography [CT], CT-myelography, or MRI), site of lesion, distribution of disc material within the cervical vertebral canal, type of disc herniation (i.e., Hansen Type I, Hansen Type II, or DAWS), surgical site, hospitalization period, intra- and postoperative complications, and long-term neurologic outcome. Based on the dogs’ body weights, dogs were categorized as either small breed (<15kg) or large breed (>15kg). Dogs were included if they had been followed-up for at least 6 mo postsurgically and if long-term follow-up information could be obtained by either neurologic re-examinations or telephone information provided by either the owner or treating veterinarian. One dog that was euthanized for complications associated with the procedure was included in the study population, as well as dogs that died or were euthanized during the study.
Neurologic Grading
Dogs were clinically graded into four different groups based on the severity of their neurologic dysfunction. Grade 0 corresponded to a normal neurologic status. Grade 1 was used to describe dogs with cervical hyperesthesia without lameness or neurologic deficits. Grade 2 corresponded to dogs with lameness and/or ambulatory tetraparesis, and grade 3 included dogs with nonambulatory tetraparesis or tetraplegia with intact nociception. Grade 4 was used to describe tetraplegic dogs with absent nociception. For all dogs, the neurologic status was graded preoperatively and at the last follow-up examination. For the purpose of this study, dogs were re-evaluated by a neurologic examination or by telephone interview with the owners, referring veterinarian, or both and assigned a final outcome assessment.
Diagnostic Imaging
Cervical spine survey radiographs were performed in all dogs. Three different imaging modalities were used to identify the lesion site: myelography, CT-myelography, and MRI. Myelography was performed by injection of Iotrolana (0.2 mL/kg) into either the cerebellomedullary or lumbar subarachnoid space. Lateral views of the cervical spine were obtained with the dog in a neutral position and in traction (i.e., by pulling the jaw and pelvic limbs in opposite directions).
CT-myelography was performed after injection of Iotrolana (0.05 mL/kg diluted in 0.25 mL/kg of physiologic saline solution) into the cerebellomedullary or lumbar subarachnoid space. Dogs were positioned in dorsal recumbency and sagittal and axial images were reconstructed after the images were acquired.
MRI was performed using a 1.5 Tesla scanner with a spine array coil. The MRI's were obtained with the dogs positioned in dorsal recumbency. Examined spinal cord area ranged from C1 to T3, and standard T2-weighted and T1-weighted fast spin echo images were acquired in the sagittal and transverse planes over the affected segment.
Localization and Classification of Spinal Cord Compression
Information obtained from diagnostic imaging and surgery reports was used to determine the site of spinal cord compression and to classify the type of IVDD as either Hansen Type I or Hansen Type II. In addition, it was noted if the lesion was associated with DAWS. DAWS was defined as Hansen Type II disc disease at C5–C6 or C6–C7, which was associated with mild vertebral malformations characterized by shortening and misalignment with displacement of the cranial portion of the segment dorsally into the spinal canal. Dogs with other causes of cervical spondylomyelopathy (e.g., hypertrophy of the ligamentum flavum or articular facets, osseus stenosis of the cervical canal) were not included.12 Whenever possible, the specific location of the herniated disc material within the spinal canal was noted after imaging diagnosis (ventral, ventrolateral, lateral). In dogs that underwent myelography or CT-myelography, Hansen Type I disc disease was suspected when there was hypoattenuating material within the intervertebral disc space or the vertebral canal, and when there was thinning and dorsal or lateral deviation of contrast columns at the affected intervertebral disc space. Hansen Type II was suspected when there was marked focal, dorsal, or dorsolateral deviation and mild thinning of the contrast columns at the affected disc spaces. In dogs that underwent MRI, Hansen Type I disc disease was diagnosed when either the disc had herniated through all of the layers of the annulus and appeared as a focal epidural mass near the affected disc space or if extruded disc material was no longer in contact with the affected disc space, but was spread out along the epidural space. Hansen Type II disc disease was diagnosed if either the disc protruded in a central, right, or left direction, sometimes with focal disruption of the annulus or if there was symmetric uniform extension of the outer disc margin circumferentially. Signal intensity of extruded and protruded disc material was hypointense in T2-weighted images in most dogs with the exception of one dog with Hansen Type I disc extrusion. In this dog, parts of the extruded disc had few foci of T2 hyperintensity, which was interpreted as fresh hemorrhage that occurred during extrusion of the disc.
Definite type of disc herniation and specific localization were determined intraoperatively. Herniation was classified as Hansen Type I based on the presence of either free sequesters of annulus fibrosus or mineralized nucleus within the epidural canal and/or under the dorsal longitudinal ligament. Hansen Type II herniations were those with protruding fibers of the annulus fibrosus without presence of sequestrated material in the epidural space.
Surgical Technique and Postoperative Care
A dorsolateral approach was used for hemilaminectomy in all dogs. Briefly, the animals were positioned in sternal recumbency and rotated away from the surgeon at an approximate angle of 45° to ease visualization of the articular processes and vertebral pedicle. The skin was incised 1–2 cm lateral to the midline at the approximate location of the lesion over a length of three vertebral segments. The underlying platysma muscle was incised to expose either the brachiocephalicus muscle in the cranial cervical spine or the trapezius muscle in the caudal cervical spine. For lesions located in the cranial cervical spine, the brachiocephalic muscle was incised parallel to its fibers to expose the underlying splenius and serratus ventralis muscles. Blunt division and retraction of these muscles revealed the underlying longissimus capitis and cervicis muscles. For lesions in the caudal cervical segments, the brachiocephalic and trapezius muscles were separated and spread in a craniodorsal and caudoventral direction to expose the splenius and serratus ventralis muscles. The scapula was abducted and caudally retracted to allow access to the C7–T1 segment. Identification of the affected segment was achieved by palpation of prominent anatomic landmarks, including the wings of the atlas, the transverse process of C6, the first rib, and/or the dorsal spinous processes of T1 and C7. From these points, the desired articular facets were palpated by counting the articular facets in either a caudal or rostral direction. Tendinous attachments of multifidus and complexus muscles of the desired articular process were incised using sharp dissection. After identification of the articular space between cranial and caudal facets, the joint capsule was incised. Large portions of the facets were removed with rongeurs, and a high speed air drill was used to create the hemilaminectomy/facetectomy. After drilling, the hemilaminectomy was completed with small rongeurs for fine bone removal. The nerve root, spinal ganglion, and lateral aspect of the spinal cord were visualized. Ventrally located disc material was removed with a blunt nerve root retractor or dental scraper (Hansen Type I). In one case with a Hansen type II disc protrusion, the dorsal annulus was incised with a scalpel no. 11 and annular fibers were excised with fine rongeurs until the spinal nerve/spinal cord returned to the floor of the spinal canal. If present, hemorrhage during removal of disc material was controlled with a macerated piece of muscle or absorbing hemostatic material. After the decompression was complete, the wound was flushed with sterile saline, and the muscles were apposed with simple interrupted sutures. The fascial layer was repaired using absorbable sutures in a simple interrupted pattern. The subcutaneous tissue and skin were closed in layers.
Postoperative care consisted of administration of either parenteral opioid analgesics or nonsteroidal anti-inflammatory drugs. Strict cage rest, avoidance of stairs, and the use of cervical collars or harnesses for at least 4 wk were recommended. Management of nonambulatory patients included nursing care and physical therapy (passive range of motion, supported gait exercises).
Outcome
Outcome was assessed by comparing the presurgical neurologic status with either follow-up examination findings (n=24) or information obtained by a telephone interview with the owner and/or treating veterinarian (n=17). The minimum follow-up time was 6 mo. The preoperative status and the neurologic grade at the last follow-up examination following surgery were compared to determine final outcome.
Treatment was considered to be effective when cervical hyperesthesia resolved, the dog regained the ability to ambulate, or if the dog improved by at least one grade in neurologic status. Outcome was considered excellent if the neurologic re-evaluation or owners’ description judged the dog to be clinically normal. Outcome was determined to be good if neurologic re-evaluation or owners’ description stated the dog was free of cervical hyperesthesia, was ambulatory, and had improved by at least one neurologic grade. If the patient was considered to have the same neurologic status before and after surgery, outcome was defined as not improved. Dogs that improved after surgery but later developed pain or neurologic deficits localized to the cervical spine during the follow-up period were considered to have had a recurrence.
Statistical Analysis
Statistical analysis was performed using nonparametric tests because of the small number of dogs enrolled in the study and the nonnormal distribution of the numerical variables. All analyses were performed using a commercially available statistical software packageb. The Fisher exact test was used to examine the significance of the association between categorized outcome and IVDD type (I and II), as well as outcome and dog size (large and small breed dogs). The χ2 test (linear-by-linear association test) was used to evaluate if there was a relationship between the operating site and outcome. The Mann-Whitney test was used to evaluate a possible relationship between the length of hospitalization and both the weight of the dog and the type of disc. A P<0.05 was considered significant.
Results
Signalment
A total of 41 dogs were included in this study. Eighteen different breeds were represented. Of the included dogs, 36% (15/41) were small, chondrodystrophic dogs (representing 9 different breeds). The dachshund accounted for 40% of the small breed dogs. The remaining 26 dogs (64%) were larger, nonchondrodystrophic dogs, with 9 different breeds represented (Table 1). Dalmatians were the most common large breed dogs, accounting for 31% (8/26) of all the large dogs.
C, cervical vertebra; CT, computed tomography; DAWS, disc-associated wobbler syndrome; F, female; FS, female, spayed; IVDD I, Hansen Type I intervertebral disc disease; IVDD II, Hansen Type II intervertebral disc disease; M, male; MC, male, castrated; No., number; T, thoracic vertebra
Thirty-nine percent (16/41) of all dogs were female. The median age was 8 yr (range, 3–13 yr). For the small dogs, the median age was 7 yr (range, 3–13 yr). The median age for the larger dogs was 8 yr (range, 4–12 yr).
Clinical and Diagnostic Imaging Findings
All dogs had cervical hyperesthesia, and in three dogs, this was the only clinical finding. Of the remaining 38 dogs, 27 (66%) had grade 2 deficits, 9 (22%) had grade 3 deficits, and 2 (5%) were tetraplegic with absent nociception (grade 4).
Myelography was performed in 28 dogs (68%) and CT-myelography was performed in 4 dogs (10%). Nine dogs (22%) underwent MRI. Overall, the C6–C7 intervertebral disc space was the most affected site, noted in 14/41 (34%) dogs (4 with Hansen Type I and 10 with Type II; in 5 dogs, there was also concurrent DAWS). The second most affected site was the C5–C6 segment, noted in 8/41 (20%) dogs (4 with Hansen Type I and 4 with Type II). The third most affected intervertebral disc space was C2–C3, noted in 7 (17%) dogs (6 with Hansen Type I and 1 with Type II). The C7–T1 segment was affected in 5 (12%) dogs (4 with Hansen Type I and 1 with Type II; Table 1). The small breed dog most affected by Hansen Type I was the dachshund (6/15, 42%). In the large breed dogs, Hansen Type I was predominantly seen in Dalmatians (5/26, 19%), and Hansen Type II was primarily noted in Doberman pinschers (5/26, 19%). Small breed dogs were most commonly affected by Hansen Type I at C2–C3 (40%, 6/15), and large breed dogs had mostly Hansen Type II at C6–C7 (42%, 11/26).
The location of spinal cord compression was divided in 6 groups: ventral Hansen Type I (n=12, 29%); ventral Hansen Type II (n=10, 24%); ventrolateral Hansen Type I (n=4, 10%); ventrolateral Hansen Type II (n=2, 5%); lateral Hansen Type I (n=6, 15%); and lateral Hansen Type II (n=7, 17%). Five Doberman pinschers and one Dalmatian with Hansen Type II disc protrusions at C6–C7 were classified as suffering from DAWS. Traction-responsive cervical disc lesions (i.e., spinal cord compressions caused by protruding annulus fibrosus) were found at C6–C7 in two Doberman pinschers (Hansen Type II), one giant schnauzer (Hansen Type II), and one Yorkshire terrier (Hansen Type I). Although Hansen Type I disc disease is usually not traction-responsive, mineralized and sequestrated disc material was found under the overlying dorsal longitudinal ligament at the time of surgery. During traction, stretching of the ligament was thought to have pushed the disc back into the intervertebral space, thereby resulting in traction responsiveness.
Surgery, Complications, and Outcome
All 41 dogs underwent hemilaminectomy. Thirty-eight (93%) had only the hemilaminectomy, but two dogs (one dachshund affected at C5–C6 and one Doberman pinscher affected at C6–C7) also underwent ventral slot decompression 1 wk and 4 wk, respectively, before the hemilaminectomy. Hemilaminectomy in these two dogs was performed because neither improved following the first surgical therapy, and follow-up imaging revealed the presence of residual disc material in the vertebral canal. One dog (Great Dane affected at C6–C7) had a hemilaminectomy 6 wk after a ventral stabilizing procedure because of persistent, severe tetraparesis due to incomplete relief of the spinal cord compression as evidenced by a follow-up CT-myelography. Approximate total surgery time was 1.5 hr for each dog.
Hemorrhage was most frequently observed during removal of herniated disc material from the floor or intervertebral foramen due to injury of the internal vertebral venous complex (25 dogs; 61%). Hemorrhage can result in the premature termination of the procedure before all of the disc material has been removed and can cause life-threatening blood loss. Neither of these hazards occurred in any of the dogs in the present report. Respiratory paralysis occurred in the early postoperative period of one dog (a rottweiler with Hansen Type I at C3–C4, preoperative grade 4). Ventilation was discontinued after 12 hr because of failure to regain spontaneous respiration.
The median hospitalization period of all dogs was 4 days (range, 0–35 days). The median hospitalization period of the large breed dogs was 5 days (range, 0–35 days), which was significantly longer than the hospitalization time of small breed dogs (median=2 days; range, 1–7 days; P<0.003). There was also a trend toward a shorter median hospitalization time in dogs with surgically confirmed Hansen Type I disc (median=3 days; range, 0–35 days; P=0.047) compared with dogs with a surgically confirmed Type II disc (median=5 days; range, 1–14 days).
Postoperative status (1 day after surgery) showed that all dogs had cervical spinal hyperesthesia, which varied in intensity but resolved over the following days. Final outcome revealed persistent neck pain in two dogs. Immediate postoperative neurologic status was improved or remained the same in all dogs, with the exception of two large breed dogs that were nonambulatory in the early postoperative period compared with their ambulatory status prior to surgery. Both dogs regained ambulatory status following surgery on days 3 and 4 postoperatively. The final outcome in these dogs was considered not improved compared with the presurgical status.
Recurrences were noted in two cases. A Doberman pinscher (affected at C6–C7 with concomitant DAWS and ventrolateral disc protrusion) showed a relapse in neck pain 24 mo after surgery. Follow-up myelography revealed new lesions at two other sites, cranial to the operated C6–C7 site. A slight, but persisting, protrusion at the operated segment was also observed. The dog was managed with nonsteroidal anti-inflammatory drugs, and was free from cervical hyperesthesia, according to a telephone interview with the owners 4 mo after initiating treatment. A giant schnauzer (affected at C6–C7 with a ventrolateral disc protrusion) showed recurrence of tetraparesis after jumping 3 mo postoperatively. Plain radiographs of the neck revealed a narrowed intervertebral space at C6-C7 without misalignment. The dog responded well to a short course of prednisone (1 mg/kg/day per os for 1 wk) and restricted activity for 10 days. That dog's final outcome 12 mo following surgery was considered good.
Follow-up myelography was available for the Bernese mountain dog. The initial diagnostic imaging findings identified a ventral disc protrusion at C6–C7, and intramedullary spinal cord hyperintensity was noted on T2-weighted images. Six months postsurgically, the Bernese mountain dog had persisting gait abnormalities (grade 2) and was considered not improved at that time. Twelve months following surgery, the owners reported increased difficulty walking with the pelvic limbs. Neurologic re-examination revealed stable neurologic function. Orthopedic and radiographic examination under anesthesia identified bilateral coxarthrosis. While anesthetized, a follow-up myelography was performed. A persistent disc protrusion at C6–C7 with moderate compression of the spinal cord was diagnosed. The combination of neurologic deficits and coxarthrosis was considered to affect the dog's quality of life, and it was euthanized by the referring veterinarian 4 wk after the repeat myelography because the dog was not responding adequately to nonsteroidal anti-inflammatory drugs.
Median follow-up time was 12 mo (range, 6–48 mo). Overall, 27/41 dogs (66%) had an excellent outcome. Specifically, 13/26 large breed dogs and 14/15 small breed dogs had excellent outcomes. In 9/41, a good outcome was noted. Eight of these dogs were large breed dogs and one was a small breed dog. Postsurgically, neurologic status was not improved in 5/41 (12%) large breed dogs. One of these large breed dogs had complications during recovery time from anesthesia and was euthanized after 24h of mechanical ventilation.
Concerning the outcome of the 22 dogs with Hansen Type I disc disease, 18 (82%) cases had an excellent outcome, including 6 large breed dogs and 12 small breed dogs. A good outcome was noted in 3 (14%) dogs, including 1 small breed and 2 large breed dogs. The 1 dog (4%) that did not improve in this group was the rottweiler mentioned above that was euthanized.
The outcome of 19 dogs with Hansen Type II disc disease was excellent in 9 (47%) cases, including 7 large breed dogs and 2 small breed dogs. A good outcome was observed in 6 (32%) of the large breed dogs, but 4 (21%) large breed dogs did not improve.
The Fisher exact test showed a significant association between excellent outcome and IVDD type (P=0.026). A linear association between outcome and dog size (P=0.007) was also noted. Specifically, the outcome in small dogs and in dogs with Hansen Type I disc disease was significantly better than in large dogs.
Discussion
Results of the current study demonstrate that cervical hemilaminectomy was effective in 88% of dogs with Hansen Type I and Type II disc herniations. Previous reports documented the benefit of this procedure for laterally located lesions. More recently, Tanaka et al. (2005) have used it successfully in six small breed dogs for removal of “symmetrically” located disc lesions.8–10 Based on the present study, the indications for cervical hemilaminectomy can be expanded to large breed dogs and dogs with ventrally and ventrolaterally located disc material for both Hansen Type I and Type II.
Previous studies reported that outcome after ventral slot decompression in small breed dogs appeared to be better for lesions in the cranial cervical region than for those with lesions in the caudal cervical region.13,14 In the current study, no difference in outcome between different areas of the cervical spine in small breed dogs with Hansen Type I disc extrusions was noted. Also, this study found an excellent outcome in large breed dogs with Hansen Type I disc extrusions, regardless of the affected segment. The current study results are in agreement with the study of Cherrone et al. (2004) who reported a successful outcome in 99% of dogs with Hansen Type I disc herniation after either a ventral slot decompression or dorsal laminectomy.15
In contrast, dogs with Hansen Type II disc protrusions tended to have less favorable outcomes in the current study. All dogs that did not improve had ventrally located disc compressions. Again, this finding is in agreement with previous investigations, which included dogs treated with ventral slot decompression for Hansen Type II disc disease and caudal cervical spondylomyelopathy (i.e., DAWS). Success rates in these studies were between 66% and 80% following surgery.16–18 In the current study, the lower success rates in dogs with Hansen Type II disc protrusions could be attributable to the difficulties in accessing ventrally located disc material (because the annulus has to be incised and the material is removed without visual control). Indeed, incomplete removal of a disc lesion of that type and location has been detected during follow-up myelography in the Bernese mountain dog. Alternatively, intramedullary spinal cord lesions after chronic compression that occur in many cases of Hansen Type II disc protrusions and DAWS may be irreversible, even if decompression was complete.19,20
The outcome of dogs with DAWS concurs with the results of Rossmeisl et al. (2005) who reported effective management of these cases in their series.9 The small number of cases with DAWS in the current study did not allow for comparisons with the outcomes of the dogs with Hansen Type II disc protrusion alone, but may not differ largely among these variants of caudal cervical disc disease.
The most frequent intraoperative complication during cervical hemilaminectomy was hemorrhage from injury to the vertebral sinuses. This complication occurred as a consequence of iatrogenic damage to the vessel or during removal of disc material from the ventral floor of the spinal canal or intervertebral foramen. Hemorrhage was controlled with local topical hemostatic agents. None of the cases reported herein had sufficient hemorrhage to necessitate cessation of the operation. Life-threatening blood loss has been reported in another study using this approach, and in the authors’ opinion, hemorrhage is the most dangerous complication associated with this surgery.8 Access to blood transfusion supplies should be available when performing this surgery.
Another noteworthy complication was the respiratory compromise in one rottweiler with preoperative grade 4 deficits due to a disc extrusion at C3–C4 after a road accident. This complication was probably associated with intraoperative spinal cord manipulation, which was likely to have compromised phrenic nerve function due to a combination of loss of upper motor neuron influence and damage to the cranial aspect of the lower motor neuron origin of the phrenic nerve. Respiratory failure has been described to occur in other decompressive procedures in the cervical area.21
Immediate postoperative deterioration of neurologic status was transiently observed in two large dogs with chronic Hansen Type II disc protrusion at C6–C7 in a ventral midline position. Reasons for deterioration could have included surgical trauma to the spinal cord during removal of disc material and/or reperfusion injury.16 Decline in neurologic status immediately following surgery has been described in dogs with this type of lesion treated by either dorsal laminectomy or ventral slot decompression and is therefore not related specifically to cervical hemilaminectomy.22
Recurrences were observed in two dogs in the late postoperative period. Myelography revealed adjacent segment disease in one dog, which is known to occur following other types of cervical surgery and was not considered as being related specifically to the hemilaminectomy.5 In the other dog, plain radiographs did not reveal any abnormalities other than a narrowed intervertebral disc space at the site of the surgery. Persistent disc protrusion, possibly aggravated by the jump that preceded the episode, or a new lesion were considered as likely causes.
Iatrogenic vertebral instability has been reported to occur in 8% of a large series of dogs following ventral slot decompression. This represents a serious complication, requiring surgical correction and subsequent stabilization.23 This complication has not been previously described for cervical hemilaminectomies and was not observed in the dogs presented herein. This suggests that cervical hemilaminectomy may have a low risk of causing clinically relevant iatrogenic destabilization of the spinal segment. This assumption is strengthened by the observation that two dogs included in the current study had a cervical hemilaminectomy performed subsequent to a ventral slot procedure without any negative clinical consequences. A combination of the ventral slot and hemilaminectomy at the same segment almost certainly decreases stiffness and, in the authors’ opinion, should be performed with caution. An accurate assessment of surgery-induced instability would require cadaveric studies comparing biomechanical behavior of operated cervical specimens by hemilaminectomy or ventral slot.
The mean hospitalization period of all dogs was 5.3 days, similar to that described in a previous report.8 Large breed dogs were hospitalized for a significantly longer period of time than small breed dogs. In a previous report studying postoperative recovery in a population of mainly large breed dogs after dorsal laminectomy, mean length of hospitalization was 8.3 days.22 The longer recovery period of large breed dogs affected with Hansen Type II disc disease might be related to a heavier body weight and the difficulties associated with removing ventral midline disc protrusions. Another possible explanation is that extensive muscle manipulation and dissection in the necks of larger dogs produces more clinically relevant tissue trauma and edema, resulting in delayed recovery.22
Limitations of this study include an insufficiently long follow-up time (i.e., it is not possible to exclude recurrences after the final outcome was determined at the time of data collection). Additionally, outcomes that were not based on neurologic re-examinations but rather on telephone interviews with owners and veterinarians are prone to information bias. Other limitations include the number of dogs in each category of disease and the retrospective nature of the study.
Conclusion
This study demonstrates that various forms of cervical IVDD from C2–T1, including Hansen Type I, Hansen Type II, and DAWS, can be treated effectively with cervical hemilaminectomy. Success rates and hospitalization times are similar to other decompressive procedures used to treat cervical disc disease. Success rates are significantly lower for large breed dogs with Hansen Type II disc lesions compared with small breed dogs with Hansen Type I disc lesions. Complications are similar to other procedures with the exception of iatrogenic destabilization. If the surgeon is familiar with the surgical technique, cervical hemilaminectomy represents an expansion of the neurosurgical repertoire in the treatment of IVDD and can be considered an appropriate method of treatment with outcomes comparable to other decompressive techniques.
Contributor Notes
