Clinical Evaluation of Rabies Virus Meningoencephalomyelitis in a Dog

Case Report

An approximately 6-month-old, 8.8-kg, intact female, mixed-breed dog was referred to the University of Florida Veterinary Medical Teaching Hospital (VMTH) with a 3-day history of acute paraplegia. The puppy was rescued from an abandoned house 2.5 weeks earlier, in a malnourished condition. The dog had been taken to the referring veterinarian twice for general examination, and the first two distemper, adenovirus, parvovirus, parainfluenza combination vaccines were administered 2 weeks apart. Rabies vaccine had not been administered at that time because of the nutritional state of the dog.

The dog initially presented to the referring veterinarian for left pelvic limb paralysis that developed over a 12-hour period. Areflexia, hyperesthesia, and nonpainful swelling of the left second and third digits of the affected limb were noted. Results of a complete blood count (CBC) and serum biochemical profile were within reference ranges, except for a mild anemia (hematocrit [HCT], 27%; reference range, 37% to 55%). Radiographs of the left foot showed only soft-tissue changes consistent with swelling of the second and third digits. Cephalexin (28.4 mg/kg body weight, per os [PO] q 8 hours) was dispensed for a presumptive bacterial cellulitis of the digits. The next day the dog was again presented to the referring veterinarian for progressive flaccid paralysis of the right pelvic limb. Abdominal radiographs were obtained and were unremarkable, with the exception of a mild dilatation of small intestines. The patient received fipronil (0.67 mL of 9.7% topical solution) for possible early tick paralysis, a warm-water enema for possible constipation, and intravenous (IV) lactated Ringer’s solution (rate unknown). Cephalexin was continued. Three days after the onset of clinical signs, the patient was referred to University of Florida VMTH.

On presentation, the animal was bright, alert, and demonstrating normal mentation. No external wounds or ticks were found. The second and third digits on the left pelvic limb were swollen, and the left popliteal lymph node was enlarged. The remainder of the physical examination was essentially normal. On neurological examination, cranial nerves were normal. Cervical flexion, extension, and palpation were normal. The thoracic limbs were questionably stiff upon wheel barrowing, but the thoracic limb spinal reflexes were normal. The pelvic limbs were paralyzed with absent patellar, cranial tibial, gastrocnemius, sciatic, and flexor reflexes. The dog was still able to wag her tail, and anal tone was normal. Deep pain sensation was normal, but a significant hyperesthesia was present over the entire surface of both pelvic limbs. Moderate focal discomfort was elicited with digital palpation over the fourth lumbar (L₄) vertebra. A bilateral lesion of the L₄ to second sacral (S₂) spinal cord and nerve roots or alteration of the neuromuscular junction was suspected. The differential diagnosis for a myelopathy included distemper meningomyelitis, spinal lymphosarcoma or other juvenile neoplasia, or rabies meningomyelitis. Differential diagnosis for neuromuscular junction blockade included acute polyradiculoneuritis (e.g., coonhound paralysis), coral snake envenomation, tick paralysis, or botulism. The initial low suspicion of rabies meningomyelitis resulted in minimal precautionary measures that permitted the following diagnostics but resulted in human exposure that could have otherwise been prevented.

Initial diagnostic tests included CBC, serum biochemical analysis, urinalysis, and abdominal radiographs (which included the thoracolumbar vertebral column), and abdominal ultrasound. Clinicopathological abnormalities consisted of a persistently low HCT of 32% (reference range, 37% to 55%), slightly decreased creatinine (0.60 mg/dL; reference range, 0.70 to 1.60 mg/dL), slightly elevated phosphorus (6.4 mg/dL; reference range, 2.1 to 6.0 mg/dL), and decreased sodium (133 mmol/L; reference range, 147 to 158 mmol/L) and chloride (96 mmol/L; reference range, 107 to 120 mmol/L). The low HCT was not further evaluated with a reticulocyte count or fecal examination. Urinalysis was within reference ranges. No radiographic abnormalities were noted in the thoracolumbar vertebral column. A persistently dilated loop of small bowel without evidence of obstruction was found on abdominal radiographs. Repeated evaluation after withholding food and water showed no obvious alteration of the small bowel dilatation. Abdominal ultrasound was unremarkable. The cephalexin was continued twice daily. No additional medications were started.

The following day, the patient was anesthetized for cerebrospinal fluid (CSF) collection, electromyography (EMG), and myelography. Analysis of lumbar CSF revealed a protein of 366 mg/dL (reference range, 0 to 31 mg/dL), red blood cell (RBC) count of 300 cells/μL (reference range, 0 RBC without blood contamination), and white blood cell (WBC) count of 1,140 cells/μL (reference range, 0 to 6 WBC/μL). Seventy-eight percent lymphocytes, 21% mononuclear phagocytes, and 1% neutrophils were noted on CSF cytopathology [Figure 1]. In addition to serum distemper and rabies virus antibody titers, serum titers were also submitted for rickettsial, Neospora caninum, and Toxoplasma gondii antibodies; CSF distemper antibody titers were also submitted. Serum rapid fluorescent focus inhibition test (RFFIT) for rabies virus was <1:5, which was considered negative. Serum distemper titer was >1:6250 for immunoglobulin G (IgG) and negative for immunoglobulin M (IgM); CSF distemper IgG titer was >1:1250 and negative for IgM. These results were consistent with the recent distemper vaccination and blood-brain barrier leakage of serum into the CSF. All other serological titers were negative.

Moderate fibrillations and positive sharp waves were found on EMG of the left pelvic limb, which were suggestive of denervation or myopathy. The right pelvic limb, para-vertebral muscles, and thoracic limb muscles had a normal EMG. No M wave could be generated on the left sciatic nerve, which suggested a lack of axonal or neuromuscular junction transmission. The motor nerve conduction velocities (MNCV) of the right sciatic and tibial nerves were 65.7 m/second and 54.4 m/second, respectively (reference range, 50 to 75 m/second). F waves were noted on the right sciatic, tibial, and ulnar nerves, but they were absent on the left. These results were suggestive of a neuromuscular junction blockade or myopathy. Due to the severe CSF lymphocytosis and the suspicion of inflammatory disease, a myelogram was not performed.

Recovery from anesthesia was slow. During recovery, the dog became demented, salivated profusely, and developed a bilateral ventrolateral strabismus. Within 8 hours of anesthesia, the dog began having focal facial and limb seizures, excessive salivation, and aggression. The dog was recovered in the isolation unit. Due to the high suspicion of rabies, the dog was euthanized and submitted for necropsy; euthanasia was 6 days after the onset of left pelvic-limb paralysis.

Half of the brain was submitted to the Florida State Diagnostic Laboratory and was positive for raccoon rabies virus by direct fluorescent antibody testing. Histopathological examination revealed a nonsuppurative moderate to severe meningoencephalomyelitis and radiculoneuritis. Rare intra-cytoplasmic inclusion bodies were seen in the brain stem and spinal cord, and a substantial number of degenerate and necrotic neurons were present within examined sections of thoracic and lumbar spinal cord [Figure 2]. There was a mild lymphoplasmacytic neuritis with fibrosis and axonal degeneration within the left sciatic nerve. At necropsy, the dilated intestinal loop visualized on radiographs was identified as jejunum; however, no histopathological cause for the dilatation could be identified.

Discussion

Rabies meningoencephalomyelitis is a rapidly fatal disease; therefore, minimal diagnostic information is available in the literature. This case did not display any of the classic signs of encephalitis at presentation; therefore, neurodiagnostics were performed to eliminate other causes of disease. Complete blood count, serum biochemical analysis, and urinalysis results did not specifically aid in the diagnosis. The decreased creatinine and elevated phosphorus were consistent with the patient’s immature status and poor nutritional state. The decreased sodium and chloride probably represented a dilution effect caused by prior IV fluid administration. The HCT was slightly decreased compared to the expected value of a 6-month-old puppy.¹ In retrospect, a reticulocyte count may have been helpful to determine the erythrocytic regenerative capacity of the patient.

A lymphocytic pleocytosis in the CSF is often associated with canine distemper and other nonspecific viral infections.^2–5 Rabies encephalitis can present with a normal or abnormal CSF cell count.²³ Although rabies and distemper were considered the most likely possibilities in this patient, titers were also run for rickettsia, Neospora caninum, and Toxoplasma gondii for potential secondary infections.

In a study of 22 cases of naturally occurring rabies in horses over a 20-year period, six horses had CSF analysis performed.⁶ Five of the six had slightly elevated leukocytes, and four of six had a predominance of lymphocytes. In another study of a colony of raccoons with rabies, CSF analysis revealed a leukocyte count >356 cells/μL in all animals with a primarily lymphocytic pleocytosis.⁷ The raccoon CSF was also evaluated for rabies virus antibodies; 100% of the animals were negative. This is hypothesized to be due to the rapid course of infection in experimentally infected animals.

The positive sharp waves, fibrillation potentials, and lack of M wave in the left pelvic limb on EMG were compatible with disease of the sciatic nerve and supported by the lymphocytic infiltration and axonal degeneration of the sciatic nerve on histopathology. The absent M wave could also have been due to the tropism of the virus for acetylcholine receptors causing a block of neuromuscular transmission.⁸ In the authors’ experience, the EMG abnormalities confined to one pelvic limb in a dog with bilateral flaccid paralysis is an unusual finding for both focal spinal cord and diffuse neuromuscular disorders. The combination of CSF and EMG abnormalities increased the suspicion of rabies meningoencephalomyelitis in this case.

In retrospect, the swollen digits of the left pelvic limb could have been secondary to an animal bite. Although rabies virus infection is a prevalent concern in wildlife throughout the United States, successful vaccination programs have greatly reduced the incidence of infection in the country’s domestic animals. In the year 2000, 7,364 cases of rabies in nonhuman animals were reported to the Centers for Disease Control and Prevention (CDC).⁹¹⁰ This number decreased from 7,961 cases in 1998. The wildlife population remained the major reservoir for rabies virus in both 1998 and the year 2000.⁹¹⁰ Prior to acquisition, the dog was tied up outdoors and left unsupervised, and this may have led to her exposure.

Rabies virus infection is typically a rapidly fatal disease with an ability to avoid immune detection during viral replication.^3–5 Low-level viral replication at the site of infection does not initiate a vigorous immune response, which may explain the negative serum RFFIT result in this case. A recently reported nucleic-acid sequence detection method for rabies virus on serum and CSF was accurate in each of four clinically affected patients.¹¹ While not all cases examined were positive on both serum and CSF, probably owing to the isolated nature of the virus, this test may become beneficial in future clinical cases.

Rabies virus travels in a retrograde manner through peripheral nerves to the central nervous system (CNS) where massive replication occurs, most notably in the spinal cord, medulla oblongata, periaquaductal gray matter, limbic system, and cerebellum.^12–14 Replication in the limbic system has been associated with the clinical manifestation of aggression reported with rabies. The virus is distributed to multiple organs, including salivary glands, through axoplasmic transport, again avoiding exposure to the immune system. The incubation time in dogs has been reported between 3 and 8 weeks, with a range of 10 days to 8.5 months.^2–515 Reports have shown that animals can shed rabies virus in saliva up to 13 days prior to onset of clinical signs.³⁴¹⁶ Most animals progress to death within 10 days due to respiratory failure.¹² Given the unknown inoculation time in this patient, it was advised that all humans and animals she contacted beginning 13 days prior to clinical signs receive medical attention. All humans involved with the patient were advised to contact a physician for additional prophylaxis.

Since rabies is fatal in humans, the public health implications of rabies are significant. Many students, staff, and faculty in the intensive care unit and neurology, radiology, and anesthesiology services were exposed during the first 24 hours of hospitalization because of the low suspicion of rabies. A total of 34 individuals at the VMTH, previously vaccinated for rabies, were required to obtain two booster immunizations due to significant contact with the puppy. The economic and emotional issues surrounding the exposure were significant given the high mortality rate of this virus. No morbidity or mortality was noted to result from rabies virus exposure. Future cases of unvaccinated pets with acute-onset flaccid paralysis should be quarantined from all personnel for 10 days or until a definitive diagnosis can be obtained.

Conclusion

Although uncommon, rabies meningoencephalomyelitis should be suspected in any unvaccinated dog with progressive neurological signs. The lymphocytic pleocytosis in the CSF and lack of M wave on electrodiagnostics further raised the index of suspicion for rabies meningoencephalomyelitis. There is not an ideal antemortem diagnostic test for rabies virus currently available. Advances in corneal impression,¹⁷ polymerase chain reaction (PCR),¹⁸ serological assays,¹¹ and CSF evaluation may provide rapid antemortem diagnostic tests in the future. To limit human exposure, all suspected cases should be maintained in an isolation area and observed for seizures, dementia, or aggression for 10 days after the onset of neurological signs. This was the first case of natural-occurring rabies in a dog presented to the neurology service at the University of Florida VMTH since its inception in 1977.