A 7-year-old 55.6-kg (122.3-lb) castrated male Great Dane was evaluated at the North Carolina State University's Veterinary Teaching Hospital because of a 2-month history of fecal incontinence. The owner speculated that the incontinence was associated with the trauma of being mounted by another Great Dane in the household; these traumatic events occurred regularly during play. The fecal incontinence appeared to wax and wane because the dog was fecally continent for periods of 1 to 2 weeks.
The owner had noticed 2 different types of incontinence. First, although the dog could defecate voluntarily when taken outside, at times it appeared unaware that it was defecating as it walked. In addition, formed stools were found on its bedding and were frequently eliminated as the dog rose from a lying position. However, the owner also reported that the dog occasionally became acutely aware of the need to defecate, but was unable to prevent defecation as it tried to get outside of the house. On these occasions, the owner thought that the dog was aware that it was defecating, but could not control the reflex. The owner was able to minimize episodes of fecal incontinence by taking the dog outside frequently. The owner had not noticed any urinary incontinence and reported that the dog postured normally to urinate and defecate, producing normal streams of urine as well as feces that were formed and normal in appearance.
Additional historical information included a long-term history of intermittent scuffing of the left hind limb. The owner had noted that the nails of the left hind foot were more worn on the dorsal aspect than those of the other feet. The dog also had a long-term history of hip dysplasia, for which it received medical treatment. One and a half months prior to the initial evaluation, treatment with glucosamine (9 mg/kg [4.1 mg/lb], PO, q 12 h) and chondroitin (7.2 mg/kg [3.3 mg/lb], PO, q 12 h) provided in a supplementa had been initiated for chronic hip dysplasia and presumed arthritis. The owner noticed mild improvement in the dog's scuffing during treatment with this medication, particularly at night. No other problems were reported. The vaccination status of the dog was current, and it was treated monthly with selamectinb (6.5 mg/kg [3 mg/lb], PO) as a flea, tick, and heartworm preventative medication. Results of analyses of blood samples performed by the referring veterinarian 6 months previously were within reference limits.
On physical examination, the dog was alert but apprehensive and panting; rectal temperature was 38.8°C (101.8°F) and heart rate was 132 beats/min. Mild muscle atrophy was evident in both hind limbs. Findings of the physical examination were otherwise unremarkable, and no orthopedic abnormalities were detected. Results of a digital examination per rectum were also considered within normal limits. On neurologic examination, the dog had an abnormal posture at rest with a wide-based hind limb stance. When walking, the dog was mildly paraparetic with moderate hind limb ataxia; the left hind limb was more severely affected than the right. Postural reactions including conscious proprioception were decreased in the left hind limb, but were considered normal in all other limbs. Mild hyperesthesia was detected via palpation of the lumbosacral junction of the vertebral column. Mentation, cranial nerve, and spinal reflexes (including the perineal reflex) were within normal limits. No further abnormalities were detected during the neurologic examination. While in the hospital, feces were found in the cage where the dog had been lying, although normal defecation was also witnessed. On the basis of the presence of paraparesis with normal spinal reflexes, the neurologic signs were localized from the third thoracic to the third lumbar spinal cord segments.
Routine blood analyses were performed. A CBC revealed mild thrombocytopenia (183 × 103 platelets/μL; reference range, 206 to 378 × 103 platelets/μL), and results of serum biochemical analyses were unremarkable. Because of the age of the dog, 3 orthogonal radiographic views of the thorax were obtained to assess for metastatic neoplasia. Radiography revealed spondylosis deformans of the thoracic portion of the vertebral column, but no other abnormalities. Abdominal ultrasonography did not reveal any notable abnormalities.
Magnetic resonance imaging of the thoracolumbar portion of the vertebral column was performed by use of a 1.5-Tesla MRI unit.c Multiple imaging sequences of the vertebral column from the third thoracic to the third sacral vertebrae were obtained, before and after IV administration of contrast mediumd; these included T2-weighted sagittal and transverse sequences, a sagittal STIR sequence, a T2-weighted sagittal myelographic sequence, and sagittal and transverse T1-weighted sequences obtained before and after administration of contrast medium. The sagittal T2-weighted and STIR images revealed subtle narrowing of the T13-L1 intervertebral space, and hypointensity of the corresponding intervertebral disk (compared with surrounding disks) that was consistent with desiccation of the nucleus pulposus at this site (Figure 1). On sagittal and transverse T2-weighted images, an area of signal void that was overlaid dorsally by an epidural mass (which was isointense, compared with the surrounding musculature) was detected in the vertebral canal above the T13-L1 intervertebral disk space (Figures 1 and 2). On transverse images, the epidural mass was located on the right side of the vertebral canal, causing displacement and compression of the spinal cord over the disk space, and extended caudally over the body of the first lumbar vertebra for approximately 1 cm. The intensity varied along the length of the mass. Compared with the spinal cord, the mass was both isointense to mildly hypointense and moderately hyperintense on T2-weighted images; on T1-weighted images, the epidural mass appeared variably hyperintense and isointense. Additionally, contrast enhancement of the periphery of the epidural material was detected on comparison of the T1-weighted images obtained before and after contrast medium administration. A small area of increased signal was evident within the spinal cord on transverse T2-weighted images; this area surrounded the central canal at the level of compression. Multiple desiccated intervertebral disks were also present in the thoracic portion of the vertebral column.
Sagittal magnetic resonance images of the thoracolumbar portion of the vertebral column of a dog that was evaluated because of a 2-month history of fecal incontinence. A—T2-weighted image. Notice the relative hypointensity of the nucleus pulposus of the T13-L1 intervertebral disk, compared with surrounding ones. There is a signal void directly over the disk space (arrow) representing herniated disk material; this is overlaid dorsally by a mass that is isointense, compared with the musculature, and hypointense, compared with the spinal cord. B—T1-weighted image obtained after IV administration of contrast medium. Notice that there is contrast enhancement of the periphery of the mass identified in panel A and that the body of the mass is hyperintense, compared with the spinal cord.
Citation: Journal of the American Veterinary Medical Association 228, 2; 10.2460/javma.228.2.230
Transverse magnetic resonance images of the vertebral column of the dog in Figure 1 obtained at the level of the T13-L1 intervertebral space. A—T2-weighted image. Notice the isointense to mildly hypointense epidural mass on the right side of the vertebral canal that is causing spinal cord compression at this site (arrow). B—T2-weighted image. Extruded disk material is evident as an area of signal void on the ventral aspect of the vertebral canal (arrow). C—T1-weighted transverse image obtained at the same level as the image in panel A, before IV administration of contrast medium. Notice that the mass (arrow) is hyperintense, compared with the spinal cord. D—T1-weighted transverse image obtained at the same level as the image in panel A, after IV administration of contrast medium. Notice that the hyperintense mass is contrast enhanced, particularly at the periphery (arrow).
Citation: Journal of the American Veterinary Medical Association 228, 2; 10.2460/javma.228.2.230
On the basis of findings of MRI, a presumptive diagnosis of a compressive intervertebral disk extrusion with secondary hemorrhage and epidural hematoma formation was made. It was speculated that extruded disk material was a component of the mass because of the region of signal void overlying the disk space at that site and because of the hypointensity of the nucleus pulposus of the intervertebral disk, which was indicative of desiccation. However, the variable appearance of the epidural compressive mass and, specifically, the increased signal on the T1-weighted images was not consistent with herniated disk material, but rather more consistent with a vascular event. The compressive lesion was therefore attributed to an organizing chronic hematoma containing primarily methemoglobin, with associated extruded disk material and spinal cord compression. Because of the contrast enhancement of the lesion in images obtained after administration of contrast medium, an epidural hemorrhagic neoplasm could not be excluded, but was considered unlikely. The area of hyperintensity within the spinal cord was presumed to be intraparenchymal edema at that site.
A right-sided hemilaminectomy was performed (centered at the T13-L1 intervertebral space) to further characterize the lesion and decompress the spinal cord. The dog was premedicated with hydromorphonee (0.05 mg/kg [0.023 mg/lb], IM) and acepromazinef (0.01 mg/kg [0.005 mg/lb], IM). Anesthesia was induced with propofolg (2.6 mg/kg [1.18 mg/lb], IV) and midazolamh (0.2 mg/kg [0.09 mg/lb], IV) and maintained via inhalation of isoflurane and oxygen. A continuous rate infusion of lidocainei (0.03 to 0.07 mg/kg/min [0.01 to 0.03 mg/lb/min], IV) was also administered during the surgery; cefazolinj (22 mg/kg [10 mg/lb], IV) was administered perioperatively. A routine surgical technique was used to perform the right-sided hemilaminectomy. A small amount of extruded disk material was found overlying the T13-L1 intervertebral space along with moderate amounts of what appeared to be an organized hematoma. The hematoma and disk material were removed and submitted for histologic analysis. The spinal cord was fully decompressed and appeared normal at the time of surgical closure. The dog tolerated anesthesia well and recovered without complications. Postoperative care included pain management with a fentanyl patchk (100 μg/h), hydromorphonee (0.05 mg/kg, IV, q 4 to 6 h as needed), and carprofenb (2.2 mg/kg [1 mg/lb], PO, q 12 h), and controlled exercise.
Histologic examination of the material obtained at surgery revealed multiple areas of pale, amphophilic matrix with scattered chondrocytes and areas of mineralization, consistent with disk material. Most of the sections analyzed contained a moderate amount of hemorrhage along with clusters of organizing proliferative fibroblasts and infiltrates of macrophages containing hemosiderin pigment. The histopathologic diagnosis was extruded intervertebral disk material with chronic hemorrhage and inflammation.
One day after surgery, the dog was able to ambulate independently, although the hind limb ataxia was slightly more pronounced than it had been previously. The dog was discharged to its owner the following day. At home, the owner had been instructed to restrict the dog's activity (ie, running, jumping, and traversing stairs) and to provide controlled exercise and supportive and incision care. The owner reported that the dog's paraparesis and hind limb ataxia had improved and had returned to the preoperative level within 1 week after discharge. Three weeks after surgery, there was complete resolution of the dog's incontinence and moderate improvements in its hind limb function. On neurologic examination, the dog retained a wide-based stance of the hind limbs as well as mild hind limb ataxia, predominantly on the left side. Postural reactions were improved, compared with findings before surgery, although the dog still had mild proprioceptive deficits in the left hind limb. Mild hind limb muscle atrophy remained. All other aspects of the neurologic and physical examinations were within normal limits. During a follow-up conversation with the owner 43 days after surgery, it was reported that the dog retained complete fecal and urinary continence, and the hind limb ataxia was unchanged.
Discussion
Fecal incontinence is a serious complication of spinal cord injuries that is typically a result of damage to the sacral spinal cord segments. However, fecal incontinence can be a problem in ambulatory dogs with injuries to the thoracolumbar portion of the spinal cord as well. Regardless of the etiology, this type of incontinence can result in euthanasia of the patient. Despite its important implications, there are few reports1,2,l of upper motor neuron fecal incontinence in ambulatory dogs with spinal cord disease, and its pathophysiologic basis is poorly understood. In the dog of this report, a chronic epidural hematoma associated with a thoracolumbar disk herniation caused fecal incontinence, which the owner considered to be a major problem.
Epidural hemorrhage as a result of tearing of venous sinuses during herniation of disk material is common and can be extensive.3–5 Organized hematomas that result in spinal cord compression are more unusual, although there are reports6,7 of epidural hematomas causing spinal cord compression in a large-breed dog and a Sumatran tiger. The MRI appearance of hematomas is reported to be dependent on several factors; included among these is the duration of a hematoma, which influences its biochemical makeup and, in turn, determines its intensity on specific MRI sequences.6,8,9 The lesion in the dog of this report was most consistent with a subacute to chronic hematoma; such hematomas (containing primarily extracellular methemoglobin) would be expected to appear hyperintense on T1- and T2-weighted images,8 compared with the spinal cord, as detected in the dog of this report. The formation of methemoglobin in such a lesion is a result of decreasing oxygen tension within the hematoma as it ages, causing deoxyhemoglobin to be oxidized to methemoglobin.9 This timing fits with the dog's history of hind limb ataxia. The mixed appearance of the lesion was hypothesized to represent repeated small hemorrhages associated with continued disruption of the intervertebral venous plexuses at this site.3,5,6,10,11 These components, consisting mainly of deoxyhemoglobin within erythrocytes, would be expected to appear isointense to slightly hypointense on T2-weighted images and isointense to hypointense on T1-weighted images, compared with the spinal cord.8 Lastly, contrast enhancement detected after IV administration of contrast medium was considered to represent intralesional neovascularization. Fecal incontinence and constipation are common findings in humans with thoracolumbar spinal cord injuries.12 As such, the mechanisms underlying incontinence have been well described in experimental models of spinal cord injury and in humans,13–17 although there is little published information regarding fecal continence associated with spinal cord injury in dogs.
Clinical characteristics of fecal incontinence in humans with upper motor neuron spinal cord lesions correlate with those detected in affected dogs. Compared with age- and gender-matched controls, affected humans with spinal cord injury (regardless of location within the vertebral column and severity) report an increased urgency to defecate as well as considerably more severe fecal incontinence.12 Among humans with upper motor neuron spinal cord injuries, once the urge to defecate was noticed, they either had difficulty maintaining continence for more than a few minutes or experienced constipation.12 Constipation is not an apparent clinical characteristic of upper motor neuron fecal incontinence in dogs, but difficulty in controlling the urge to defecate appears to be a major problem in dogs as described in other studies1,2,l and in the case reported here.
The mechanisms underlying these clinical findings are manifold. Continence is normally maintained by resting EAS tone, mediated through the Onuf's nucleus and its corresponding influence on lower motor neurons such as the pudendal nerve,13 as well as voluntary EAS contraction against challenges such as rectal and colonic distention, rectal contraction, and increased abdominal pressure.12–14 Additionally, the rectoanal excitatory and inhibitory reflexes induce EAS contraction and IAS relaxation, respectively, in the face of loading in the distal portion of the colon.13–15 The sympathetic nervous system is the major regulator of basal anal pressure via IAS contraction as well as a potent inhibitor of colonic motility.14,16 These mechanisms, coordinated via cortical influence, result in synchronized relaxation of rectal smooth muscle and contraction of pelvic striated musculature to prevent defecation.14
Reflex activities remain intact in humans and other animals with spinal cord injuries in the thoracolumbar region. However, lack of upper motor neuron modulation of these mechanisms in those patients results in uncoordinated, inefficient contractions and episodic incontinence.14 Specifically, incontinence in patients with spinal cord injury is related to decreased rectal sensation, loss of conscious EAS control, and colonic hypercontractility and lower basal anal pressure (IAS tone) in response to low distending volumes.14–17 The latter is the result of the lack of tonic influence from the sympathetic nervous system.16 These factors overall result in a loss of sensation of rectal distention and an exaggerated tendency toward defecation.14 In ambulatory humans with spinal cord injury, this is evident clinically as the inability to withhold defecation for normal periods as well as incontinence related to decreased awareness of defecation,13 as detected in the dog of this report.
On consideration of the nature of the fecal incontinence, the presence of a lesion in the thoracolumbar portion of the spinal cord, and the resolution of incontinence following decompressive surgery, the authors hypothesized that the fecal incontinence in the dog of this report was a consequence of the thoracolumbar spinal cord injury. The dog's normal perineal reflex and anal tone, along with absence of abnormalities in images of the lumbosacral portion of the vertebral column, ruled out the more common lower motor neuron incontinence.
In the veterinary literature, there are reports1,2,l of 2 different types of spinal cord disease in ambulatory dogs that are associated with upper motor neuron fecal incontinence. One type of disease involves cystic spinal cord lesions such as subarachnoid cysts.1,l Skeen et al1 reported cases of fecal and urinary incontinence associated with spinal arachnoid cysts in the cervical and thoracic portions of the vertebral column in dogs.1 In those dogs, fecal incontinence developed prior to the owners detecting hind limb abnormalities, similar to the dog of this report. All dogs in that study1 retained a normal perineal reflex and normal anal tone; incontinence improved in 2 of those dogs after marsupialization or fenestration of the cysts.
The other type of disease has been identified in dogs with long-term fecal incontinence after recovery from severe injury to the thoracolumbar portion of the spinal cord.2 Affected dogs lost deep pain perception and hind limb motor function as a result of severe thoracolumbar spinal cord injuries caused by trauma or intervertebral disk extrusion or herniation. They subsequently recovered sensation and motor function, but approximately 40% of them continued to have difficulties with fecal continence. Similar to the dog of this report, those dogs appeared to be aware of their need to defecate but were not able to wait until they were taken outdoors. Likewise, defecation in the house was more frequently associated with excitement and with being left indoors for long periods, and sometimes dogs did not appear to be aware that they were defecating. None of the affected dogs had signs of lower motor neuron incontinence.2 The nature of the fecal incontinence in these more severely injured dogs was similar to that in the dog of this report, but the former were paraplegic at the time of injury.2
Despite clinical similarities in the nature of the fecal incontinence between the dog of this report and the aforementioned types of spinal cord disease in ambulatory dogs that are associated with upper motor neuron fecal incontinence, the former is unusual in that there was no evidence of cystic spinal cord disease in that dog and the fecal incontinence was the primary problem for which the owner sought veterinary medical attention. Although the dog did have neurologic deficits in the hind limbs, the owner did not perceive them to be a major problem and indeed had not noticed them until they were pointed out. It was also unusual that the fecal incontinence completely resolved following surgical decompression of the spinal cord. It is important to recognize that upper motor neuron fecal incontinence can result from a variety of thoracolumbar spinal cord lesions in dogs and may be responsive to decompressive surgery. As illustrated by the dog of this report, epidural hematomas can result from disk herniations and become large enough to cause clinically important compression of the spinal cord.
| MRI | Magnetic resonance imaging |
| STIR | Short T1 inversion recovery |
| EAS | External anal sphincter |
| IAS | Internal anal sphincter |
Nutramax Laboratories, Edgewood, Md.
Pfizer Animal Health, LaJolla, Calif.
MAGNETOM Symphony MRI unit, Siemens Medical Solutions USA Inc, Malvern, Pa.
Gadoversetamide, OptiMARK injection, Mallinckrodt Imaging, St Louis, Mo.
Baxter Health Care Co, Deerfield, Ill.
Vetus Animal Health, Rockville Centre, NY.
Scherring-Plough Corp, Irvine, Calif.
American Pharmaceutical Partners Inc, Schaumburg, Ill.
Butler Co, Dublin, Ohio.
Watson Laboratories Inc, Corona, Calif.
Janssen Pharmaceuticals, Titusville, NJ.
Chen AV, Bagley RS, West CL, et al. Fecal incontinence as a primary presenting complaint in 5 dogs with cystic abnormalities of the spinal cord (abstr). J Vet Intern Med 2004;18, 408.
References
- 1↑
Skeen TM, Olby NJ, Munana KR, et al.Spinal arachnoid cysts in 17 dogs. J Am Anim Hosp Assoc 2003; 39: 271–282.
- 2↑
Olby NJ, Levine J, Harris T, et al.Long term functional outcome of dogs with severe injuries of the thoracolumbar spinal cord: 87 cases (1996–2001). J Am Vet Med Assoc 2003; 222: 762–769.
- 3
Hoerlein BF. Intervertebral disc protrusions in the dog. I. Incidence and pathological lesions. Am J Vet Res 1953; 14: 260–269.
- 4
Shores A. Intervertebral disc syndrome in the dog. Part I. Pathophysiology and management. Compend Contin Educ Pract Vet 1981; 3: 639–645.
- 5
Olby NJ, Munana KR, Sharp NJH, et al.The computed tomographic appearance of acute thoracolumbar intervertebral disc herniations in dogs. Vet Radiol Ultrasound 2000; 41: 396–402.
- 6
Tidwell AS, Specht A, Blaeser L, et al.Magnetic resonance imaging features of extradural hematomas associated with intervertebral disc herniation in a dog. Vet Radiol Ultrasound 2002; 43: 319–324.
- 7
Ketz-Riely CJ, Galloway DS, Hoover JP, et al.Paresis secondary to an extradural hematoma in a Sumatran tiger (Panthera tigris sumatrae). J Zoo Wildl Med 2004; 35: 208–215.
- 8↑
Platt SR, Garosi LG. Canine cerebrovascular disease: do dogs have strokes? J Am Anim Hosp Assoc 2003;39:337–342.
- 9↑
Thomas WB, Adams WH, McGavin MD, et al.Magnetic resonance imaging appearance of intracranial hemorrhage secondary to cerebral vascular malformation in a dog. Vet Radiol Ultrasound 1997; 38: 371–375.
- 10
Gomez M, Freeman L, Jones J, et al.Computed tomographic anatomy of the canine cervical vertebral venous system. Vet Radiol Ultrasound 2004; 45: 29–37.
- 11
Jaramillo MG, Freeman L. Review of the vertebral venous plexus in the dog. Int J Morphol 2003; 21: 237–244.
- 12↑
Lynch AC, Wong C, Anthony A, et al.Bowel dysfunction following spinal cord injury: a description of bowel function in a spinal-cord injured population and comparison with age and gender matched controls. Spinal Cord 2000; 38: 717–723.
- 13↑
Sun WM, Read NW, Donnely TC. Anorectal function in incontinent patients with cerebrospinal disease. Gastroenterology 1990; 99: 1372–1379.
- 14↑
Holmes GM, Rogers RC, Bresnahan JC, et al.External anal sphincter hyperreflexia following spinal transection in the rat. J Neurotrauma 1998; 15: 451–457.
- 15
Lynch AC, Anthony A, Dobbs BR, et al.Anorectal physiology following spinal cord injury. Spinal Cord 2000; 38: 573–580.
- 16↑
Frenckner B, Ihre T. Influence of autonomic nerves on the internal anal sphincter in man. Gut 1976; 17: 306–312.
- 17
Hulten L. External nervous control of colonic motility and blood flow. An experimental study in the cat. Acta Physiol Scand Suppl 1969; 335: 1–116.
