Pathologic and immunohistochemical findings in goshawks (Accipiter gentilis) and great horned owls (Bubo virginianus) naturally infected with West Nile virus.

The carcasses of 25 great horned owls and 12 goshawks were investigated for West Nile virus (WNV) infection by immunohistochemistry (IHC) performed on various organs, including brain, spinal cord, heart, kidney, eye, bone marrow, spleen, liver, lungs, pancreas, intestine, and proventriculus, using a WNV-antigen-specific monoclonal antibody and by WNV-specific reverse transcriptase-polymerase chain reaction (RT-PCR), performed on fresh brain tissue only. WNV infection was diagnosed by IHC in all owls and all goshawks. WNV-specific RT-PCR amplified WNV-RNA in the brain of all goshawks but only 12 owls (48%). Cachexia was a common macroscopic finding associated with WNV infection in owls (76%). Myocarditis was occasionally macroscopically evident in goshawks (33%). Microscopically, inflammatory lesions, including lymphoplasmacytic and histiocytic encephalitis, myocarditis, endophthalmitis, and pancreatitis were present in both species but were more common and more severe in goshawks than in owls. The most characteristic brain lesion in owls was the formation of glial nodules, in particular in the molecular layer of the cerebellum, while encephalitis affecting the periventricular parenchyma of the cerebral cortex was common in the goshawks. In owls, WNV-antigen-positive cells were present usually only in very small numbers per organ. Kidney (80%), heart (39%), and cerebellum (37%) were the organs that most commonly contained WNV antigen in owls. WNV antigen was frequently widely distributed in the organs of infected goshawks, with increased amounts of WNV antigen in the heart and the cerebrum. Spleen (75%), cerebellum (66%), heart (58%), cerebrum (58%), and eye (50%) were often WNV-antigen positive in goshawks. In contrast with the goshawks, WNV antigen was not present in cerebral and retinal neurons of owls. WNV infection appears to be capable of causing fatal disease in great horned owls and goshawks. However, the distribution and severity of histologic lesions, the antigen distribution in the various organs, and the amount of antigen varied among both species. Therefore, the diagnostician may choose organs for histology and immunohistochemistry as well as RT-PCR depending on the investigated species in order to avoid false-negative results.