Introduction

Baylisascaris procyonis, the common raccoon roundworm, is a highly fatal zoonotic disease that has been rarely reported in dogs. Between 1973 and 2002, 46% of human cases with known outcomes were fatal.¹ Nearly 70% of all adult raccoons and 90% of all juvenile raccoons are infected by B. procyonis, with the highest prevalence occurring in the Midwest, Northwest, and West Coast of the United States.^2–4 The roundworm is routinely identified in bedding such as hay or straw and in the feces of infected animals in regions populated by raccoons.^1,5 Canine infection with B. procyonis can occur secondary to consuming infective eggs in the environment or from ingesting stage three larvae within the tissues of rodent prey, which serve as intermediate hosts.^6,7 As with other roundworm species, B. procyonis undergoes the following three main forms of somatic migration within aberrant and paratenic hosts: visceral larva migrans, ocular larva migrans, and neural larva migrans (NLM).⁶ Studies in animal models suggest that ∼5–7% of ingested B. procyonis eggs migrate to the central nervous system.⁶ NLM secondary to a natural infection with B. procyonis has been rarely reported in dogs.^8,9 This report describes the clinical presentation, progression, and necropsy findings in a young dog diagnosed with NLM secondary to Baylisascaris procyonis.

Case Report

A 16 wk old, 8.9 kg intact female Gordon setter dog was presented to the University of Wisconsin-Madison neurology service for further evaluation of a 2 wk history of difficulty walking. She was acquired from a breeder at 9 wk old and had been healthy other than a recent diagnosis of Coccidia, treated with sulfadimethoxine (dose unknown). She was housed outdoors, in a rural area of Wisconsin, and slept in a pen containing straw.

Vaccination for distemper virus had been performed thrice; however, the rabies virus vaccination had not been administered to date.

Two weeks prior to presentation, the dog was evaluated by an emergency veterinarian for acute onset ataxia. A complete blood count and serum chemistry were performed, showing a mild leukocytosis (18.9 × 10³ cells/μL; reference 5.0–14.0 cells/μL), mild neutrophilia (12.59 × 10³ cells/μL; reference 2.6–10.0 cells/μL), and moderate eosinophilia (2.659 × 10³ cells/μL; reference 0.1–1.7 cells/μL). Serum biochemistry was within expected parameters. An infectious etiology was suspected therefore clindamycin (17 mg/kg per os [PO] q 12 hr) was initiated. The clinical signs progressed and prednisone (0.8 mg/kg PO q 12 hr) was added 3 days after starting clindamycin.

After no improvement, the dog was referred to the University of Wisconsin-Madison neurology service for further evaluation. Physical examination abnormalities were limited to the neurologic system. On neurologic examination, the dog had the following abnormalities: an intention tremor, truncal sway, hypermetria most prominent in the thoracic limbs, circling left, mild left head tilt, positional ventrolateral strabismus left eye, bilateral miosis with intact pupillary light reflex, and absent postural reactions of the right thoracic limb. All reflexes, remaining cranial nerves, other postural reactions, and spinal palpation were unremarkable or normal.

Neuroanatomic lesion was localized to the cerebellovestibular system, more prominent on the right side but with a bilateral component. The differential diagnoses included infectious or primary inflammatory meningoencephalitis, neoplasia or neurodegenerative disease such as a storage disorder. Recommended diagnostic tests included infectious disease titers, thoracic radiographs, brain MRI, and cerebrospinal fluid (CSF) analysis. Infectious disease titers for Cryptococcus neoformans, Neospora caninum, and Blastomyces dermatitidis were negative^ab. Virus reverse transcription polymerase chain reaction for distemper was negative^c. All remaining diagnostic testing was declined. Prednisone and clindamycin were continued at previous dosages while awaiting results from infectious disease testing, and meclizine was added (1.4 mg/kg [3.0 mg/lb] PO q 24 hr).

As a result of lack of improvement, the dog was euthanized 1 wk after neurologic examination, and a full necropsy was performed. There were no apparent gross lesions of the central nervous system. The cerebellar size was grossly proportionate to the cerebrum, with no evidence of cerebellar herniation. On histopathologic examination, the most striking finding was an extensive focus of necrosis in the white matter of the cerebellum where the architecture was effaced by sheets of swollen macrophages with intracytoplasmic globular eosinophilic material (cellular breakdown products). Adjacent to this focus was an ∼150 μm granuloma composed of epithelioid macrophages concentrically arranged around a vaguely linear aggregate of eosinophilic material. Special stains looking for fungal organisms (Grocott-Gomori’s methenamine silver stain) fortuitously highlighted three cross sections of a 75 μm wide nematode larva with a 5 μm cuticle, prominent lateral cords, lateral alae, and coelomyelian musculature. (Figure 1) Attempts to section organisms elsewhere in the brain were unsuccessful. Other white matter degenerative changes, characterized by multifocal axonal swelling with intraluminal macrophages and occasional spheroids, were randomly noted in the cerebellum and in the temporal and parietal lobes (Figure 2). Overall, the findings were consistent with multifocal degeneration of the white matter in both the cerebellum and cerebrum with low numbers of migrating nematode parasites associated with a focus of cerebellar necrosis. When comparing with other nematodes, the relatively large size and morphology of the prominent lateral alae was most consistent with B. procyonis.

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

Discussion

This case highlights a rare but important zoonotic disease that should be considered as a differential diagnosis in any dog exhibiting acute, progressive central nervous system signs and peripheral eosinophilia. Exposure to B. procyonis can occur through contact with contaminated bedding or through consumption of intermediate hosts. The highest prevalence of B. procyonis in raccoons is found in the Midwest, Northwest, and West Coast of the United States.^2–4

After being ingested, eggs hatch in the small intestine, penetrate the intestinal walls, and migrate through the liver to the lungs via portal circulation. Subsequently, via pulmonary veins, they gain access to systemic arterial circulation, thereby gaining access to one of the three main areas of somatic migration.⁶ Within the central nervous system, larval migration can produce severe tissue trauma and inflammation. Changes may be seen between 3 days and 2 wk after infection.^3,6,10,11

Clinical signs of NLM depend on the number of larvae that migrate and the extent of infection.⁴ Acute onset ataxia, progressing to recumbence within 48 hr, was reported in one dog naturally infected with B. procyonis. The possible exposure was traced to a raccoon housed in close proximity to the affected dog.⁸ The dog in the current report demonstrated a longer clinical course than previous reported, which may reflect the parasite load, treatment administered, or host immunity. Behavioral changes, including irritability and decreased activity, reflex abnormalities, and hypertonia have been noted in children diagnosed with B. procyonis. Depending on the extent and severity of NLM, these signs can progress to opisthotonus, stupor, coma, seizures, and death.^1,3 Between 1973 and 2002, 46% of human cases with known outcomes were fatal.¹ In the remaining cases, long-term sequelae included weakness, spasticity, developmental delay, blindness, and seizures.¹ The dog in this report exhibited signs of progressive cerebellovestibular dysfunction, which corresponded with the location of nematode migration noted on histopathologic examination.

Diagnosing B. procyonis intestinal infections antemortem poses several challenges. First, although adult dogs may shed B. procyonis eggs, they can easily be mistaken for Toxocara canis eggs. Subtle distinctions of B. procyonis eggs include a more irregular and roughened shell compared with T. canis eggs, which have a more regular, pitted surface.⁷ Additionally, adult female worms may not be present in the intestines in cases in which larval somatic migration has occurred. In addition to fecal analysis, several other techniques have been explored in cases in which B. procyonis infection is suspected in human cases. Enzyme-linked immunosorbent assays have shown promise in supporting a B. procyonis diagnosis, but one-way cross-reactivity with T. canis has been observed.¹² Western blotting can identify antigens unique to B. procyonis in the serum of patients with known B. procyonis larva migrans.¹² Testing for antibodies against B. procyonis in a serum with indirect immunofluorescence has also been performed with convalescent increases in titers in serial monitoring in one human case study.¹³ The most common finding on hematologic and CSF analysis are eosinophilia and an eosinophilic pleocytosis.^1,2 Moderate eosinophilia was documented in this dog at presentation to the emergency veterinarian; however, CSF was not obtained. Common MRI findings include nonspecific acute cerebellar or periventricular deep white matter changes.¹⁴ However, white matter changes tend to lag behind clinical signs, reducing the effectiveness for early diagnosis.^1,14 Overall, the sensitivity and specificity of these tests have not been determined in dogs.²

A definitive diagnosis of NLM is based on the presence of variable numbers of larvae, either with or without associated inflammation, on postmortem examination. In comparison with T. canis, B. procyonis larvae are larger (50–70 μm in diameter) with prominent lateral alae and paired excretory columns lateral to a central digestive canal.^1,6,8 The histopathologic examination abnormalities noted in this case are similar to findings reported in other dogs and humans diagnosed with B. procyonis.^1,8

At this time, there are no approved treatments for B. procyonis. Most drugs that are approved for treatment of T. canis are considered effective against B. procyonis.^7,15 Administration of anthelminthics between 1 and 7 days after exposure may prevent infection; however, most infected animals are subclinical during this time. In cases in which NLM has already occurred, even if larvae are killed, significant central nervous system damage may have already occurred; therefore, the prognosis for recovery even with appropriate treatment is grave.¹ Albendazole crosses the blood-brain barrier, and therefore may limit sequelae of NLM if administered as soon as clinical signs of NLM are observed. In some cases, simultaneous prednisone therapy may help reduce albendazole and larval associated inflammation.^3,16 This combination was used successfully in an 11 mo old boy, although he continued to have seizures, decreased vision, and encephalopathy on follow-up 1.5 yr after initial treatment.³ The dog described in this report received anti-inflammatory prednisone during the initial clinical phase, which may have delayed clinical progression. There are no published reports of successful recovery in dogs with B. procyonis NLM.

Conclusion

In summary, this case highlights a rare but important zoonotic disease that should be considered as a differential diagnosis in any dog exhibiting acute, progressive central nervous system signs and peripheral eosinophilia. The index of suspicion should be elevated in young dogs with an unknown deworming history and known or suspected exposure to raccoons. Given the challenges in obtaining an antemortem diagnosis of NLM, this case also demonstrates the need for a sensitive and specific screening test for B. procyonis exposure or infection. Because most anthelminthic have been shown to be effective against B. procyonis, regular deworming is recommended to reduce the likelihood of NLM and zoonosis.