Degenerative lumbosacral stenosis (DLSS) is common in working Police German Shepherds and has a complex aetiopathogenesis. DLSS is an acquired narrowing of the vertebral canal +/- intervertebral foraminae resulting in compressive radiculopathy of the cauda equina. Whilst it is accepted that degeneration of the L7–S1 IVD plays a major role, whether that degeneration is primary (due to heritable/anatomic factors) or secondary to repetitive strain is still debated. As the disc degenerates (fibroid metaplasia) it loses the ability to constrain the vertebral link. Motion increases, especially axially, adding increased stresses on the remaining components of the spinal unit (facet joints, interarcuate ligament, dorsal ligament).

A degenerate IVD cannot preserve the foraminal pathway for the L7 nerve at extreme range of motion. The sacral lamina and facet are thrust further forward into the dorsal and lateral articulation, affecting the size of the L7/S1 foramen. The foramen is not a simple aperture, but rather a tunnel, with entrance, middle and exit zones. Type II disc protrusion in a dorsolateral direction affects the entrance and middle zones of the foramen, while impingement of the exit zone results from narrowing of the IVD space, osteophytosis or hypertrophy of soft tissue in the region of the facets. Instability, defined as ventral subluxation of the sacrum, is seen in some cases but is probably an extreme situation. More common is wedging of the disc and collapse of the intervertebral spacing.

The degree of anatomical change affects the nature and severity of the clinical signs detected. Central protrusion causes dysfunction of the sacral and caudal nerves, leading to abnormal carriage of the tail and faecal and/or urinary incontinence. In contrast, lateral (foraminal) compression of the L7 nerve roots leads to nerve root signature with intermittent to progressive lameness and sciatic deficits in severe cases.

Surgical treatment has traditionally been directed at dorsal decompression of the cauda equina via a dorsal L7–S1 laminectomy. Type II disc protrusions have been treated with annulectomy or discectomy through the laminectomy. Yet experimental data shows that these two procedures reduce the spinal unit stability even further, so they may be counterproductive. Indeed, results of dorsal laminectomy without annulectomy are the same or superior to dorsal laminectomy with annulectomy or discectomy. Dorsal laminectomy poorly decompresses the L7–S1 foramen which is lateral and rostral beneath the caudal articular facet of L7. A facetectomy allows improved exposure but destabilises the spine. Lateral foramenotomy allows the cranial part of the foramen to be enlarged (Godde, Steffen 2007). An alternative approach is to perform a stabilisation procedure designed to maintain foraminal volume by limiting extension.

Presenting Signs and Physical Examination Findings

Working dogs typically present early in the course of the disease. The demands of working soon highlight a problem that might go unnoticed in a pet animal. Handlers report reluctance to jump, or pain when jumping or rising from a prone position/climbing stairs. Some dogs show exertion-induced intermittent lameness (nerve root signature), e.g., after scaling walls. In advanced cases, more typical of pet animals, there can be scuffing of the toe nails, urinary or faecal incontinence and paraparesis.

Pain on direct digital palpation of the lumbosacral area and hyperextension are the most consistent presenting clinical signs of DLSS. Other variable signs include hind-limb weakness and/or ataxia, urinary and/or faecal incontinence, and a flaccid tail. Careful palpation may reveal atrophy of the gluteal or stifle flexor muscles. Examination should include pressure applied dorsally over the lumbosacral space with and without extension of the hips, tail jack, and the 'lordosis test.' Pain elicited on extension of the hip is not specific for lumbosacral disease, as it can be associated with either spinal or coxofemoral joint pain. Pain on extension and abduction or rotation of the hip is more suggestive of coxofemoral joint pain. Neurological examination may reveal depressed cranial tibial, sciatic and withdrawal reflexes, and normal to exaggerated patellar reflexes. This 'pseudo-exaggeration' is the result of flaccidity of the flexor muscles of the stifle, which antagonise the patellar reflex and should not be confused with upper motor neuron disease.

Diagnostic Imaging

In early cases (as typical for working dogs) there are only mild plain radiographic findings. More advanced cases have ventral and/or laterodorsal spondylosis and dynamic studies may show wedging/instability. However, these changes are nonspecific and advanced imaging is required. CT and MRI are both useful and have pros and cons. MRI has been preferred over CT due to its greater soft tissue differentiation. The nerves of the caudal equina can be seen within epidural fat and absence of the fat signal is a sensitive indicator of impingement. However, high-resolution multi-slice CTs now have excellent soft tissue differentiation. In addition, CT can be easily reconstructed into 3D images allowing the foramen to be directly observed from any direction including within the canal. At the MUVTH we use a protocol developed by Angela Hartman, DACVR. We position the dog in flexion, neutral, and extension and run both soft tissue and bone algorithms. As a surgeon I like to look at the 3D images and plan the surgery based on the presence of dynamic or static foraminal impingement in either one or both foramen and the spinal canal. In early cases, there are few osteophytes and impingement is dynamic, due to buckling of dorsal soft tissue, protrusion of the disc and excessive motion of the S1 lamina/facets. These dogs benefit from stabilisation with a limited laminectomy of S1. In advanced cases, more typical in pets, spondylosis is advanced and new bone impinges on the foramen or canal. Motion has decreased in these cases and the spinal unit is fusing naturally. The nerves can be severely impinged and a foraminotomy or facetectomy (with dorsal fusion) is required.

Treatment

Conservative management with NSAIDs has been recommended for pet dogs with pain only and whose lifestyle can be modified to avoid strenuous exercise. Repeated epidural infiltration of methylprednisolone has been shown to be beneficial at reducing clinical signs in 30/38 (79%) dogs in a preliminary study, and 20 were considered by their owners to be free of signs (median followup 48 months) after a median of five injections (Janssens et al. 2009). Less than a quarter of the dogs in the above study were working/sporting breeds. At the MUVTH, we have attempted this regime in a few GSD police dogs. In most cases there was some improvement, but the dogs relapsed when back at work in the longer term. We now reserve medical management for older dogs close to retirement.

The outcome following traditional dorsal decompressive surgery worsens with age and the presence of neurological deficits (inc. urinary or faecal incontinence). The results of several studies (all breeds) using dorsal decompression and annulectomy yielded a mean total percentage cure of 55%, and improved the clinical signs in a further 25% of dogs with DLSS. Recurrence of clinical signs is reported by the owner or diagnosed by clinical examination in 18–37% of dogs. The prognosis for dorsal decompression in working dogs with DLSS is poorer than for pet animals, prompting alternative approaches.

In 1986, Slocum and Devine advanced a technique for facet joint fixation and bone grafting to permanently fixate the lumbosacral joint in a neutral position. They proposed that hypertrophy of soft tissue causing compression of neural structures was secondary to instability and would regress once the lumbosacral joint was rigidly fixated. In their technique, partially threaded pins or screws are placed through the articulation of the facet joint, and an autogenous bone graft is placed dorsally to promote spondylosis. The long-term results of dorsal fixation/fusion have only been reported in two small-scale studies (Slocum and Devine 1986; Meheust et al. 2000), with a total of 13 dogs, all reportedly having good postoperative outcomes.

Screw fracture and failure of the fixation of the screw have been observed, and concurrent dorsal laminectomy of the caudal aspect of L7 results in facet weakening, making the facet prone to fracture. For this reason, I limit the laminectomy to the cranial aspect of S1. Others have used specialised medical devices for pedicle screw fixation; however, these implants are cost prohibitive. The author is investigating the use of the 'string of pearls' (Orthomed, Huddersfield, UK) locking plate for dorsal fixation/fusion. Two 3-hole plates are used, one on each side. Each has a screw in the pedicle of L7 and the cranial aspect of the sacrum. This technique allows a more aggressive laminectomy/dorsomedial foraminotomy, with decreased concern over weakening of the facet joint. Early clinical results have been encouraging, and the implants are undergoing mechanical testing.

The lateral foramenotomy procedure represents a significant advance in the management of DLSS. In a retrospective clinical study, Godde and Steffen (Vet Surgery 2007) reported 19/20 dogs had a good to excellent outcome at a mean followup of 15 months. A small prospective clinical study in non-working dogs at the MUVTH yielded an excellent outcome in 6/8 dogs. The remaining two dogs developed worsening and more widespread neurological signs attributable to a degenerative myelopathy. Immediate postoperative CT scans in 2 dogs showed the foraminal volume to be 400% larger as the result of foramenotomy. Seven dogs had follow-up scans a median of 14 months post-op. There was evidence of some remodelling of the site with the foraminal volume reducing to 300% of the preoperative. However, the long-term success of the technique in working dogs is unknown.

References

1.  Godde T, Steffen F. Surgical treatment of lumbosacral foraminal stenosis using a lateral approach in twenty dogs with degenerative lumbosacral stenosis. Veterinary Surgery. 2007;36:705–713.

2.  Slocum B, Devine T. L7–S1 fixation-fusion for treatment of cauda equina compression in the dog. Journal of the American Veterinary Medical Association. 1986;188:31–35.

3.  Worth AJ, Thompson DJ, Hartman AC. Degenerative lumbosacral stenosis in working dogs: current concepts and review. New Zealand Vet J. 2009;57(6):319–330.