Lumbosacral intradural-extramedullary intervertebral disc extrusion in a cat

Abstract

Case summary

A 10-year-old female spayed domestic shorthair cat was evaluated for a 6-week history of abnormal tail carriage and constipation. Examination revealed tail paresis and pain over the lumbosacral and sacrocaudal articulations and on tail manipulation. MRI revealed a contrast-enhancing mass within the vertebral canal over the lumbosacral disc space, compressing the cauda equina. The mass filled the epidural space, resulting in complete attenuation of the cerebrospinal fluid signal. Laminectomy and durotomy were performed over L7–S1, revealing white, firm material within the subarachnoid space. Microscopically, the material was consistent with degenerative intervertebral disc material. Postoperatively, the clinical signs resolved completely.

Relevance and novel information

Intervertebral disc herniation (IVDH) is uncommon in cats, with most cases involving extradural compression of nervous tissue. Reports describing intramedullary intervertebral disc extrusions in cats are rare. To the authors’ knowledge, the present case is the first reported intradural-extramedullary intervertebral disc extrusion in a cat. Although MRI can often delineate extradural lesions, it can be insensitive in differentiating intradural-extramedullary from intramedullary lesions. In the present case, the location of the lesion within the vertebral canal at the lumbosacral disc space made the determination of the lesion’s location with respect to the meninges challenging. Moreover, the strong contrast enhancement of the lesion raised an index of suspicion for neoplasia. Surgical intervention and histopathology confirmed an intradural-extramedullary IVDH. The present case adds to a growing body of literature regarding IVDH in cats and details the imaging findings of intradural-extramedullary IVDH in a cat.

Case description

A 10-year-old female spayed domestic shorthair cat, weighing 4.1 kg, was presented for a 6-week history of drooped tail carriage, painful tail manipulation and constipation. Before presentation to the University of Georgia, the cat had been examined and treated with prednisolone (0.4 mg/kg PO q12h for 3 days, then 0.4 mg/kg PO q24h for 3 days, then 0.4 mg/kg PO q48h for 3 days) and gabapentin (10 mg/kg PO q12h). Initially, therapy resulted in improved tail carriage. After 3 days of treatment, the owner discontinued the medications. Abnormal tail carriage recurred. Reinstitution of prednisolone and gabapentin failed to improve clinical signs. In addition, constipation started, with the cat straining to defecate without producing stool. The cat was treated with a warm water enema and discharged with instructions to continue prednisolone and gabapentin. One week later, signs persisted. The cat was hospitalized for 2 days for intravenous fluids (lactated Ringer’s solution at 3.6 ml/kg/h) and several enemas. A complete blood count and chemistry profile were normal except for neutrophilia (12.48 × 10³/μl, reference interval 2.30–10.29). The cat was discharged with prednisolone (0.4 mg/kg PO q12h), gabapentin (10 mg/kg q8h), polyethylene glycol 3350 (¼ tsp mixed in food q12h, Miralax; Bayer Healthcare) and cisapride (0.6 mg/kg PO q8h, Propulsid; Janssen Pharmaceuticals) and the diet was changed to a high-fiber diet. Three weeks later, constipation persisted, and the cat’s tail appeared abnormally positioned and painful. Consequently, the cat was referred for further evaluation.

On referral examination, the physical examination was normal apart from tail paresis. Neurologically, the cat had a normal mentation, gait, postural reactions, spinal reflexes and cranial nerve examination. It exhibited pain on tail extension and palpation over the lumbosacral and sacrocaudal articulations. Tail nociception was intact, and both the perineal reflex and anal sphincter tone were normal. Neurological examination was consistent with lower motor neuron tail paresis, secondary to a lesion affecting the caudal spinal cord segments, caudal nerves or muscles of the tail. The differential diagnosis included intervertebral disc herniation (IVDH) or neoplasia. Infectious or inflammatory myelitis or neuritis was considered unlikely. Serum electrolytes and three-view thoracic radiographs were normal. Radiographs of the vertebral column from L3 through the tail showed mild narrowing of the lumbosacral disc space, spondylosis deformans at the L5–L6 and L7–S1 articulations, and incomplete fusion of the S2 and S3 vertebral bodies with sacrocaudal malalignment (Figure 1).

Figure 1

On the lateral radiograph, the lumbosacral intervertebral disc space is mildly narrowed with ventral spondylosis deformans (white arrow). There is incomplete fusion of S2–S3 (white arrowhead) with narrowing and mild subluxation of the sacrocaudal articulation. Ventral spondylosis at L5–L6 is also present (black arrowhead)

With the cat positioned in dorsal recumbency and the coxofemoral joints extended, 3.0-T MRI (Skyra; Siemens) of the lumbar vertebral column, sacrum and cranial tail was performed. Multiplanar T2-weighted (T2W), half-Fourier single-shot turbo spin-echo (HASTE), and pre- and postcontrast T1-weighted (T1W; 0.2 ml/kg of gadolinium-based contrast medium, Ominscan; GE Health Care) images were acquired. Postcontrast T1W images were acquired using the Dixon fat saturation technique. An ovoid mass measuring 5 × 1.75 mm was present within the vertebral canal centered over the lumbosacral intervertebral disc (IVD) space, compressing the cauda equina. The mass filled the epidural space causing complete attenuation of the cerebrospinal fluid (CSF) signal and epidural fat (Figure 2). The mass was isointense to the spinal cord on T2W and T1W images. On precontrast images, the mass was difficult to discriminate from the spinal nerves of the cauda equina and conus medullaris. The mass was right-sided and displayed homogeneous contrast enhancement (Figure 3). Meningeal enhancement of the lumbar spinal cord was present. The spatial resolution of the images was not sufficient to define the location of the lesion with respect to the meninges. Given its location over the narrowed lumbosacral IVD space, an extradural location was presumed.

Figure 2

(a) On median sagittal T2-weighted imaging, an ovoid mass (arrow) is present within the vertebral canal overlying the lumbosacral intervertebral disc space. The mass appears to fill the epidural space, resulting in attenuation of epidural fat and cerebrospinal fluid. There also is a loss of signal from the nucleus pulposus at L2–L3, L5–L6 and L7–sacrum. (b) On median sagittal T1W imaging, the mass (arrow) is less conspicuous. (c) On the median sagittal 2D-HASTE image, there is complete attenuation of the subarachnoid space at the lumbosacral articulation (arrow). (d) On a postcontrast median sagittal T1W image acquired with fat saturation, the mass displays strong homogeneous contrast enhancement (arrow). T1W = T1-weighted

Figure 3

On sequential (a–c, from cranial to caudal) postcontrast T1-weighted images acquired using the Dixon fat saturation technique, the mass (arrow) appeared as a slightly right-sided epidural lesion that displayed strong, homogeneous contrast enhancement

The cat underwent a lumbosacral laminectomy via a dorsal midline approach to the L7–S1 articulation, with the cat positioned in sternal recumbency with the coxo-femoral joints flexed. ¹ A 4–5 cm skin incision was made, and the subcutaneous fat and lumbosacral fascia were sharply incised. The multifidus lumborum and sacrocaudalis dorsalis medialis muscles were elevated. A high-speed pneumatic drill was used to create a laminectomy. Within the epidural space, the dural sac was enlarged and dorsally displaced. The sacral spinal nerves were erythematous. A 25 G hypodermic needle was used to perform a 5–6 mm craniocaudal durotomy. Despite the durotomy, CSF did not leak from the subarachnoid space. Within the subarachnoid space, abutting the end of the conus medullaris and compressing the intradural spinal nerve roots, a firm, white mass (measuring approximately 0.1 × 0.2 × 0.4 cm) was identified and removed. As the mass was removed, CSF flowed from the durotomy (see video in supplementary material). The laminectomy was covered with a free fat graft, and the surgical site was closed in a routine manner.

Histologically, the mass consisted of degenerate cartilage with areas composed of fibroblasts, blood vessels, macrophages and neutrophils, consistent with degenerate IVD material.

Postoperatively, the cat was treated in the hospital for 1 week with prednisolone (0.5 mg/kg PO q12h for 7 days, then 0.5 mg/kg PO q24h for 7 days, then 0.5 mg/kg q48h for 7 days), gabapentin (10 mg/kg PO q8–12h for 14 days) and lactulose (10 g/15 ml; 2 ml PO q8–12h as needed). Over the course of hospitalization, the cat’s tail pain resolved, and voluntary movements of the tail improved. At re-examination 10 days after discharge, the owner reported the cat had returned to normal and was defecating normally without lactulose administration. On examination, the cat was not in pain and displayed normal tail movements.

Discussion

IVDH is categorized as intervertebral disc extrusion (IVDE), where degenerative nucleus pulposus extrudes through a disrupted annulus pulposus, or as intervertebral disc protrusion (IVDP), where both annulus fibrosus and nucleus pulposus shift into the vertebral canal. ² In cats, the prevalence of IVDH is 0.44% or lower.^{3 –5} Studies conflict over the greater relative prevalence of IVDE over IVDP.^{4 –7} Intramedullary IVDH (IIVDH), in which IVD material enters the spinal cord parenchyma, has been rarely reported in cats.^{4,8 –10} To the authors’ knowledge, the present case is the first description of an intradural-extramedullary (ID-EM) IVDE in a cat. The term ID-EM IVDH defines the presence of IVD material within the subarachnoid space. Moreover, it implies that spinal cord parenchyma is present at the site where the IVD material has penetrated the dura mater. In cats, the conus medullaris is consistently located within the lumbosacral vertebral canal, unlike its more cranial location in dogs. ¹¹ The termination of the spinal cord in the present case was located at approximately the midbody of L7 on MRI. Therefore, the spinal cord parenchyma was not present at the lumbosacral disc space where the IVD material had penetrated the dura. However, in surgery, the IVD material within the subarachnoid space was visibly abutting the end of the conus medullaris. It is possible that the extended position of the coxofemoral joints during MRI compared with their flexed position during surgery altered the location of the termination of the spinal cord in relation to the vertebrae. Dynamic shifting of the termination of the spinal cord related to positioning of the coxofemoral joints does occur in dogs. ¹² In dogs, the termination of the spinal cord shifts more caudally with the coxofemoral joints in flexion. ¹² We have used the term ID-EM IVDH herein as the IVD material was identified adjacent to the caudal end of the conus medullaris during surgery. An alternative characterization would have been to describe the IVDE as intradural.

In the present case, MRI findings suggested an extradural lesion. In cats, MRI findings of extradural IVDH include loss of T2W hyperintensity of nucleus pulposus, T2W hypointense extradural material and spinal cord compression. ³ Based on the presumed extradural location of the lesion centered over the lumbosacral IVD, which displayed a dehydrated nucleus pulposus, IVDH was considered more likely than neoplasia.^5,13 However, an ID-EM lesion was identified at surgery. Although the determination of an extradural location of a lesion is not typically challenging, discriminating between ID-EM and intramedullary lesions may be difficult on MRI. ¹⁴

In dogs, MRI findings of ID-EM IVDE include the golf-tee sign, the Y sign and the beak sign. ¹⁵ Improved discrimination of the location of IVDE relative to the meninges may require specialized MRI sequences. T2*-weighted imaging has been used to confirm intramedullary intervertebral disc extrusion (IIVDE) in cats. ¹⁰ The use of three-dimensional steady state free procession (3D-SSFP) imaging and two-dimensional (2D) or 3D HASTE sequences may increase the conspicuity of ID-EM IVDE. ¹⁶ Although a 2D HASTE sequence was acquired, given the small size of the anatomy combined with the poor spatial resolution, findings consistent with an ID-EM IVDE were not appreciated; the subarachnoid space abruptly attenuated as is observed with extradural lesions. Even with 3D-SSFP or 3D-HASTE imaging, features of ID-EM pathology may not have been identified given the small size of the dural sac enveloping the conus medullaris. In dogs, the MRI characteristics of ID-EM IVDE are based on the IVDE occurring cranial to the termination of the spinal cord. In this case, the IVDE was at the level of the spinal cord termination, as seen on MRI. Hence, the typical MRI features of ID-EM IVDE described in dogs were not observed in our cat.

Our misclassification of the lesion as extradural, and therefore prioritizing IVDH, likely influenced the owner’s decision to pursue surgery given the favorable prognosis associated with IVDH in cats. Had we considered neoplasia the most likely diagnosis, the owner’s decision to pursue surgical intervention may have differed. The prognosis for extradural IVDE is reported to be excellent, with approximately 80% of cats regaining function after decompressive surgery. ⁴ Similar success rates occur with medical management; however, cats treated medically tend to have less severe neurological deficits than cats in which surgical therapy is pursued. ⁴ Moreover, the success rate for lumbosacral IVDH treated surgically is also reported to be excellent.^4,17,18 Success rates for the medical treatment of lumbosacral IVDH in cats have not been defined. Successful outcomes with medical therapy for IIVDE are also reported.^{4,8 –10} In the present case, the addition of exercise restriction, a component of medical management that is generally recommended, would likely not have resolved the clinical signs. In cats with extradural lumbosacral IVDH, pain and constipation often resolve after decompression.^4,17,18 In the present case, the constipation was likely the result of discomfort as neurological deficits were limited to tail paresis. In less successful cases, pelvic limb paresis can persist.^4,17,18 Therefore, surgical intervention was considered more likely to improve the cat’s clinical signs. The successful outcome in the present case was likely related to the limited scope of clinical signs (tail paresis, pain and constipation) and surgical decompression.

Conclusions

The decision to pursue surgery or medical therapy for cats with myelopathies hinges on the presumed etiology, which is predicated on the lesion’s anatomic location relative to the meninges. The benefits of different treatment options expressed by clinicians factor into owners’ decisions whether to pursue surgical or medical therapy. The present case illustrates the potential for a successful outcome of surgical intervention for lumbosacral ID-EM IVDE in cats. The potential for ID-EM IVDE should be considered in cats with intradural lesions on MRI. Although care should be exercised in considering surgical exploration, this case underscores that even when MRI findings are equivocal, exploratory surgery can be both diagnostic and therapeutic, and in cases of lumbosacral intradural IVDE in cats, it can lead to a full recovery.