DUVV infection was recently confirmed in a 77-year-old man with type 2 diabetes who was scratched on the face by what appears to have been an insectivorous bat in February 2006 in North West Province, South Africa, ≈80 km from the location where the first DUVV infection occurred 36 years earlier. The bat flew into a room at night, landed on the man's spectacles while he was attempting to chase it out, and scratched his face as he brushed it off. The bat did not appear to have bitten him, and it escaped after the incident. He did not seek medical care, and thus no postexposure treatment was given. He became ill at home in Cape Town 27 days later and received treatment for influenzalike illness. He slept most of the following day, had hallucinations that night, and was admitted to a hospital on the third day of illness. On admission, he had a fever (40°C), tachycardia, neck and general limb rigidity, hyperreflexia, facial fasciculation, and involuntary grimacing. Within 24 hours generalized tonic-clonic seizures had developed with status epilepticus supervening. These necessitated intubation, sedation, and mechanical ventilation. He died on day 14 of his illness.

Heminested reverse transcription–PCR was performed as previously described (9) with modified forward primer JW12 (10). This procedure detected lyssavirus nucleic acid in saliva taken on day 10 of illness and in brain tissue collected postmortem. Nucleotide sequencing of the PCR products and phylogenetic analysis performed as previously described (11) confirmed the identity of the agent as DUVV (Figure), and live virus was isolated from saliva and brain tissue by mouse inoculation. Immunofluorescence tests with antirabies conjugate prepared to be cross-reactive with the African rabies–related viruses (Onderstepoort Veterinary Institute, Pretoria, South Africa) showed small and sparse inclusion bodies in impression smears of the cortex, hippocampus, thalamus, medulla, and cerebellum. Histopathologic examination of sections from the cortex, hippocampus, thalamus, hypothalamus, midbrain, pons, medulla, and cerebellum showed polioencephalitis affecting predominantly the diencephalon and brainstem and involving varying degrees of neuronopathy, neuronal loss, astrocytosis, parenchymal and perivascular lymphocytic infiltration with CD45 immunopositivity, sparse macrophage activation, and axonal spheroid formation. No nuclear or cytoplasmic inclusions were observed.

The ability to distinguish between various lyssaviruses and monitor their relative distribution and prevalence has important implications for implementation of control measures. It was recognized in 1932 that mongoose-associated rabies in South Africa differs from classic dog rabies (2). Although an inadequately characterized lyssavirus was isolated from a bat trapped in a survey in 1963, before the existence of rabies-related viruses was known, awareness of lyssaviruses other than rabies viruses dates from the identification of DUVV in 1970 and was followed by detection of LBV and MOKV in South Africa (6,7,12). Routine differentiation of diagnostic isolates became feasible with the availability of monoclonal antibodies during the 1980s and the subsequent introduction of molecular epidemiology techniques (7,13–15). Although instances of persons seeking rabies prophylaxis after exposure to bats have been reported, the recent case of DUVV infection constitutes only the second known instance of a person in South Africa with lyssavirus infection after such an encounter. Nevertheless, it is clear that rabies-related viruses are widely endemic in South Africa and that active investigation of the bat-associated lyssaviruses is warranted.