Antemortem Diagnosis of Human Rabies

Sir—We previously reported the use of nucleic acid sequence-based amplification (NASBA) for the detection of rabies virus RNA in samples of saliva, CSF, and/or urine obtained during life from 8 patients infected with rabies [1, 2]. Here, we summarize the results of NASBA in correlation with the clinical onset of symptoms of the disease. From September 1998 through March 2004, we collected and tested 58 specimens from 23 rabies patients (20 with furious and 3 with paralytic rabies). Samples collected included: 27 saliva, 14 CSF, 15 urine, and 1 tear specimen; and extracted hairs from 1 patient. All patients had been bitten by stray dogs. Postmortem brain samples from all patients were positive for rabies virus by either fluorescence antibody testing or mouse inoculation; all samples were also positive for rabies virus by either NASBA or RT-PCR. All samples, except the first 4 specimens, which were frozen, stored, and examined retrospectively [1], were kept at 4°C for 24–48 h until examined for the presence of the rabies nucleocapsid gene. In 21 of 23 patients, we identified rabies RNA in specimens obtained on the first day of hospitalization. Specimens collected within 3 days after clinical onset yielded the highest number of positive results with saliva samples having the highest rate of positivity (7 of 8 specimens), followed by CSF (4 of 6) and urine (2 of 5). The sensitivity of all specimen types dropped after 3 days; however, saliva remained the most practical and reliable source for virus detection (11 of 15 specimens positive during days 4–6 after onset and 1 of 2 positive during days 7–9). The test sensitivity for urine (3 of 9 specimens positive for rabies virus RNA) and CSF (2 of 7) was comparable during days 4–6. Test results for 2 saliva specimens, 1 CSF specimen, and 1 urine specimen obtained during days 10–12 were all negative. Of particular interest were the test results for hairs extracted from 1 patient obtained 4 days after onset of symptoms. Fifty hair samples were extracted from this patient instead of excising skin with hair follicles from the nape of the neck [3]. We were able to demonstrate the presence of rabies RNA in the ends of the hair follicles.