Neuropathology and diagnostic features of rabies in a litter of piglets, with a brief review of the literature

Abstract

Porcine rabies is exceedingly rare worldwide. We describe herein the neuropathology and the diagnostic features of an outbreak of rabies in a litter of piglets attacked by a skunk in Georgia, United States. Rabies viral infection was confirmed in 2 of 3 piglets submitted for testing. Inflammatory and degenerative changes were more prominent in the brainstem and consisted of lymphoplasmacytic meningoencephalitis with glial nodules, neuronal necrosis, and neuronophagia. No viral inclusions (Negri bodies) were observed in multiple sections of brain. A fluorescent antibody test on fresh samples of brainstem and cerebellum was confirmatory for the eastern United States raccoon rabies virus variant. Immunoreactivity for rabies virus was detected across all brain sections in both cases but was more prominent in the thalamic and brainstem nuclei, as well as in the medial lemniscus. Rabies is an important differential diagnosis in pigs with neurologic disease.

Rabies is an often fatal zoonotic disease of mammals characterized by progressive neurologic signs resulting from viral infection of the central nervous system (CNS).² Rabies virus (Rhabdoviridae, Lyssavirus, Rabies lyssavirus) is transmitted by the saliva of an infected animal, typically through a bite.^1,18 Wild and domestic animals account for > 90% and 7.6% of the reported cases of rabies in the United States, respectively.¹² In the southeastern United States, particularly Georgia, raccoons are the most common reservoirs of the virus and thus play an important role in the spread of the disease.¹² Rabies is exceptionally rare in pigs in the United States, with only 3 cases reported in 2015 and no cases described since then,³ to our knowledge. Similarly, the overall incidence of porcine rabies is low worldwide, with rare case reports mainly from the United States, Canada, Brazil, China and other parts of Asia, and many African countries.^{6,9,15,16,18,20,21}

After primary replication within the musculature near the bite wound, viral particles reach the CNS via retrograde axonal transport, leading to a multitude of degenerative and/or inflammatory changes in the brain or spinal cord.² After a period of replication in the CNS, virions use axons to gain access to the salivary glands, from which they will be readily available to complete the transmission cycle and infect a new susceptible host.² The nature and distribution of the pathologic changes in the CNS of infected animals depend largely on the primary site of viral inoculation, viral variant, and affected species.¹⁹ Further, the distribution of viral antigen within the CNS varies according to different animal species and is an important aspect in the diagnostic confirmation of rabies, given that submission of samples with low antigen load may yield false-negative results and compromise the diagnosis.¹⁹ Given the rarity of porcine rabies, the distribution of lesions and viral antigen in the brain has not been well characterized in pigs. Herein we describe an outbreak of rabies in a litter of piglets in Georgia, focusing on the distribution of the pathologic changes and viral antigen in the brain and on the laboratory tests used for diagnostic confirmation.

In April 2018, a skunk attacked 9 piglets and a sow in an outdoor pen. Four piglets with severe cutaneous wounds were euthanized and 2 other piglets died suddenly within the following week. No human or other mammalian exposure was noted. One of the 2 dead piglets was submitted for rabies testing at the Georgia Public Health Laboratory (GPHL, Decatur, GA) and was reportedly negative. Of the 3 remaining live piglets, 1 exhibited pelvic limb paralysis, lethargy, and sternal recumbency (case 1) 20 d after exposure; the other 2 piglets (cases 2 and 3) had no clinical signs. These 3 piglets were euthanized and submitted to the Athens Veterinary Diagnostic Laboratory (AVDL; University of Georgia, Athens, GA) for rabies testing.

The brains were removed, and cross-sections of cerebellum and brainstem were submitted for fluorescent antibody testing (FAT) for rabies virus according to the AVDL procedures for rabies-suspect cases. No gross anatomic changes were present in the brains. Cases 1 and 2 were positive for rabies via FAT; case 3 was negative. Frozen samples from the positive and negative cases were sent to the GPHL for diagnostic confirmation and viral typing. FAT performed at the GPHL confirmed that cases 1 and 2 were positive for the eastern United States raccoon variant of rabies virus; case 3 was confirmed negative.

The remaining brain tissue was fixed in 10% neutral-buffered formalin. Representative tissue sections of brain (frontal, parietal, and temporal telencephalon, basal nuclei, thalamus, hippocampus, mesencephalon, pons, medulla oblongata, and cerebellum) were subsequently sectioned, routinely processed for histology, and stained with hematoxylin and eosin. All tissue sections were also subjected to immunohistochemistry (IHC) for rabies for determination of the distribution of viral antigen in the brain. Briefly, IHC was performed on an automated stainer (Nemesis 3600; Biocare Medical, Concord, CA). A goat polyclonal antibody for rabies (Light Diagnostics, Burlington, MA) at a dilution of 1:6,000 for 60 min was used. Antigen retrieval on tissue sections was achieved using citrate solution 10× (BioGenex, Fremont, CA) at a dilution of 1:10 for 15 min at 110°C. A biotinylated rabbit anti-goat antibody at a dilution of 1:300 (Vector Laboratories, Burlingame, CA) was utilized to detect the target, and immunoreaction was visualized using 3,3-diaminobenzidine (DAB; Biocare Medical) substrate for 12 min counterstained with hematoxylin. Rabies FAT–positive, bovine brain tissue was used as a control.

Histologically, moderate-to-marked lymphoplasmacytic meningoencephalitis (Fig. 1) with widespread glial nodules, neuronal necrosis, and neuronophagia (Fig. 2) was present in the thalamus, mesencephalon, pons, and medulla oblongata in cases 1 and 2. Similar but less severe changes were occasionally present throughout the telencephalon, including the frontal and parietal cerebral cortex, basal nuclei, hippocampus, and cerebellum in case 2. No eosinophilic intracytoplasmic viral inclusions (Negri bodies) were observed with routine histology in any brain section of either piglet. Immunoreactivity for rabies virus was robust in the thalamic and brainstem nuclei (Fig. 3), particularly the vestibular and facial nerve nuclei, as well as the medial lemniscus in both cases; scant immunoreactivity was observed within large pyramidal cells within the basal nuclei of case 1, where inflammation was absent. Strong immunolabeling was also detected along neuronal tracts, including cerebellar Purkinje cells, in case 2. Case 3 had no pathologic changes or IHC staining in the multiple tissue sections of brain examined.

Figure 1.

Porcine rabies, case 1. Inflammatory cells (lymphocytes and plasma cells) expand perivascular spaces (arrow) and infiltrate the adjacent thalamic neuroparenchyma (arrowheads). H&E. Figure 2. Porcine rabies, case 2. A glial nodule (center) and multiple necrotic neurons are present in the pons. H&E. Inset: a necrotic neuron is surrounded by phagocytic cells (neuronophagia). H&E.

Figure 3.

Porcine rabies, case 2. Immunolabeling for rabies viral antigen in neurons and their processes in the pons. Immunohisto-chemistry, DAB chromogen and hematoxylin counterstain.

Sporadic cases of porcine rabies have been reported worldwide, including dog- and bat-associated cases in Brazil and China (and other parts of Asia) and wildlife-associated cases in Canada and the United States.^{6–9,11,13–16,21} In areas where bat-associated rabies is endemic, affected pigs typically develop pelvic limb paralysis.¹⁶ In contrast, dog-associated rabies usually leads to aggressiveness and hyperactivity.^9,16 As we found in our cases, porcine rabies in Canada and the United States is typically caused by interactions with wildlife.^3,4,21 The incubation period for case 1 was ~16 d, which is on the upper limit of the reported incubation period of rabies in pigs caused by dog bite.¹¹ Although the skunk was not captured and tested for rabies, the incubation period determined in the 2 infected piglets and the viral variant suggest that the skunk attack was the transmission event in this outbreak.

Based on our cases and other reports from North America, wildlife-associated rabies in pigs is predominantly characterized by clinical signs of progressive paralysis.^7,8,21 These clinical signs are attributed to the overall distribution of the lesions in the CNS, which tend to predominate in the brainstem and spinal cord.¹⁶ A detailed assessment of the clinical history and distribution of pathologic changes in the CNS of rabid pigs is not well characterized in most of the reported cases.^7–9,20,21 Inflammatory and degenerative changes in our 2 cases occurred predominantly in the thalamus and brainstem, which is consistent with the clinical signs exhibited by one of the pigs and with the rare cases of porcine rabies that had detailed histologic examination of the brain.^16,19

The lymphoplasmacytic inflammation and neurodegenerative changes present in our cases are common to many other viral causes of meningoencephalitis in pigs.⁵ Based on routine histology, other differential diagnoses should be considered in case rabies could not be confirmed, including porcine teschovirus, encephalomyocarditis virus, pseudorabies virus (Suid alphaherpesvirus 1), West Nile virus, Japanese encephalitis virus, porcine reproductive and respiratory syndrome virus (PRRSV), porcine sapelovirus, and eastern equine encephalomyelitis virus infections.^5,17 An important finding in our cases was the absence of Negri bodies after detailed histologic evaluation of multiple sections of brain. Historically, the detection of these viral inclusions has been considered almost pathognomonic for the diagnosis of rabies, but their presence is likely dependent on the stage of infection and is often reported as unreliable, especially in pigs, in which inclusions are typically absent.^{10,13,15,16,19} For these reasons, a rabies diagnosis should not rely solely on the presence of these inclusions but must always be confirmed by other ancillary tests such as FAT, IHC, or PCR.

Our finding of strong immunoreactivity within the brainstem of our 2 cases is consistent with the findings in a previous report of a rabid pig.¹⁸ The preferential distribution of rabies antigen in the CNS of other domestic animal species has been described, including hippocampus of dogs and cats, brainstem and cerebellum of cattle, and cervical spinal cord and brainstem of horses.¹⁹ These specific anatomic sites are of great importance to ensure the accuracy of a diagnosis when submitting samples for diagnostic confirmation. Although reports of porcine rabies are rare, it appears that brainstem would be the recommended neuroanatomic area for viral antigen detection in pigs based on the current and previously described rabies cases.¹⁹

Follow-up with the referring veterinarian revealed that the sow, which was also attacked during the initial incident, was quarantined for 6 mo and released after not developing neurologic disease, in accordance with the 2016 post-exposure protocol for livestock.⁴