The Diagnosis of West Nile Virus Infection in Horses.

Abstract: This letter refers to the Veterinary Pathology brief communication “West Nile Virus Infection in Horses: Detection by Immunohistochemistry, In Situ Hybridization, and ELISA” by Toplu et al.5 The authors attempt to describe the clinicopathologic findings in naturally occurring West Nile virus (WNV) infection in horses. WNV was diagnosed in a foal by immunohistochemical and in situ hybridization techniques, and the presence of WNV antibodies was detected in 5 other horses with clinical signs suggestive of WNV infection. The foal was the only animal that received a complete postmortem and histological examination. The authors indicate that the clinical signs were “consistent” with West Nile virus infection. Clinical signs described are not specific since several other diseases may reveal with the same clinical presentation. In the method for immunohistochemistry they refer to a fluorescence antibody technique, but the represented image (Figure 3) is not of such technique. Instead, it depicts an indirect peroxidase DAB technique with brown staining. This immunohistochemical image consists of a cerebral blood vessel with endothelial and periendothelial staining; however, several other surrounding cells are labeled as well. West Nile flavivirus in horses is not characterized by frequent and abundant staining and is not characterized by the endothelial infection seen in avian species. In the image there is also evidence of nuclear staining although RNA flavivirus infection is not characterized by nuclear infection. In Figure 2, which represents the in situ hybridization (ISH) localization of the flavivirus nucleic acid, as also observed in the IHC viral detection image, all the cells types in the field are labeled, with extension of the staining almost diffusely into the neuropil without clear cut localization in fibers or cytoplasm. In summary, WNV IHC, and ISH stain appearance, distribution, and localization in this manuscript suggest nonspecific staining. The authors also describe that the foal is affected by lymphadenomegaly, lymphoid tissue necrosis, and hepatocellular necrosis. These are all lesions that do not occur with an equine WNV infection in horses. The presence of these lesions and the dubious IHC and ISH staining suggest that this foal was affected by another, perhaps more significant, condition and that WNV was not necessarily a significant cause of disease in this foal if present at all (no confirmatory polymerase chain reaction [PCR] was performed). Regarding the other horses, it seems that we will never know for sure either if WNV was the cause of their disease since the necropsy findings were not reported and the antibody of the competitive ELISA kit used for the detection of anti-pr-E antibodies cross-reacts with Japanese Encephalitis viruses and the tick-borne encephalitis virus. The preferred way for the postmortem diagnosis of West Nile virus infection in mammals is PCR testing of brainstem tissue along with the histologic identification of rhombencephalomyelitis with perivascular lymphohistiocytic cuffs, sparse phagocytic glial nodules, and ring hemorrhages. The intralesional viral antigen will be scant and contained in the cytoplasm of a few neurons, nerve fibers, and small aggregates of glial cells. In some cases numerous sections (sometimes more than 10) will have to be processed to detect the viral antigen. Combination of diagnostic tests is very often the best approach to a diagnosis and caution should be used in the interpretation of IHC and ISH.1–4,6