Detection of rabies viral RNA by TaqMan real-time RT-PCR using non-neural specimens from dogs infected with rabies virus.

To determine the burden of rabies in developing countries, a reliable and accurate diagnostic test for the examination of the brains of animals is needed. Recently, the number of samples and carcasses submitted to rabies diagnostic units has been declining. Methods for obtaining tissues from different regions of the brain are even more difficult, and direct florescent antibody examination may fail if the samples decomposed. The spread of rabies virus to peripheral non-nervous tissues starts early during the pre-clinical phase. It has been shown that saliva and skin biopsies taken at the neck and containing hair follicles can be used in the ante-mortem diagnosis of rabies in humans. Obtaining oral swab samples, whisker or hair follicles from the heads of canines is easy and practical and can be performed without special equipment. The objective of this study was to determine whether these non-neural specimens can be used for the detection of rabies viral RNA. The RNAs extracted from these specimens were tested using a real-time reverse transcriptase polymerase chain reaction (RT-PCR). The sensitivity of the TaqMan real-time RT-PCR analysis using samples from dogs confirmed to be infected with rabies virus was 84.6% (55/65), 81.8% (54/66) and 66.7% (44/66) when using oral swab samples, extracted whisker follicles and extracted hair follicles; the specificity of all specimen types was 100%. The negative predictive values were 77.8%, 74.4% and 61.4%, respectively. Although the rate of positivity when combining the three non-neural specimen types was increased to 86.4%, this level of sensitivity was not sufficient to help physicians whether to administer post exposure prophylaxis. However, these oral swab and whisker specimens may serve to enhance epidemiological surveillance; such data will contribute in the planning of rabies control programs.