Description of sample collection
Rabies is a notifiable disease in Bhutan and any suspected cases must be reported to the veterinary authority for investigation [11]. Following report of any rabies suspect case(s) in animals in the community, veterinarian and or laboratory technician (listed authors in this paper) conducted investigation and collected brain tissue samples (preferably brain stem, Ammon’s horn, thalamus, cerebral cortex, cerebellum and medulla oblongata) by opening the skull in their respective region. Protective personal equipment (such as hand gloves, face mask and eye google) were always worn while opening the skull and collecting the brain samples. Most of the rabies suspect stray dogs would have been already killed by the people when the team arrived at the site for investigation and sample collection while some suspect cases were euthanized to collect the samples and to confirm the case. However, the samples from other livestock were collected after the death of animals since it was culturally sensitive to euthanize the animals. The diagnostic samples in this study included both from carcasses and euthanized animals. Majority of the samples had originated from rabies endemic areas in the southern part of Bhutan that share border with India (Figure 1) while few samples were also collected from rabies-free interior Bhutan and tested as part of the routine surveillance program.
Firstly, a rapid test was performed in the field using fresh brain tissue (described below) and the second batch of samples were preserved in phosphate glycerol saline (50%) and shipped to the national veterinary laboratory located at the National Centre for Animal Health (NCAH), Thimphu for conducting FAT. Depending on the outbreak location, sample referral from the field to the national laboratory would normally take about 2-5 days.
Description of rapid immunochromatographic test (Anigen Rapid Ag test)
We used Anigen Rapid Rabies Ag Test Kit (BioNote, Inc, Hwaseong-si, Korea) in the field. The basic principle behind this test is the fluid migration of a sample along a nitrocellulose membrane. Gold conjugated antibodies are directed against epitopes of the rabies virus nucleoprotein [22, 26] and the antigen-antibody complex is then immobilized by a second antibody which is fixed on the test strip. The rapid test kit has a letter of “T” and “C” as test line and control line on the surface of the device. Both the test line and control line in result window are not visible before applying any samples. The control line is used for procedural control. Control line should be always appeared if the test procedure is performed properly and the test reagents of control line are working. A purple test line will be visible in the result window if there are enough rabies virus antigen in the specimen. The detection limit of this kit is about 102.0LD50/0.03ml in mice [12].
For this study, the rapid test was performed immediately after harvesting the fresh brain tissue samples in the field according to the manufacturer’s instruction (BioNote, Inc) (Figure 2). Briefly, the cotton swab supplied along with the kit was used to swab the brain tissue (hippocampus in case of dog and cerebellum in case of cattle, goat, pig, and sheep) and then dipped the swab into the specimen tube containing 1ml of assay diluent, and stirred/mixed well to ensure a good sample extraction. The swabbing of the brain tissue with swab and mixing into the assay diluent was repeated for 3-5 times to ensure a good sample extraction. The test cassette/device was removed from the foil pouch and placed it horizontally on a flat and dry surface. Using the disposable dropper provided with the kit, four drops of the extracted sample (100μL) was added into the sample hole in the cassette and the result was interpreted within 5-10 minutes according to the manufacturer’s instruction. As the test begin to work, the appearance of two colour bands - one on the control line (C) and the other on test line (T) within the result window, no matter which band appears first indicates a positive result. The appearance of only one colour band within the result window indicates a negative result and if the purple color band is not visible within the result window after performing the test, the result is considered invalid. The test results were immediately shared with the animal owners or any people that had linked with the exposure. All samples whether positive or negative to rapid test were submitted to the national laboratory for performing FAT (described below). At the national laboratory, the sample details and the rapid test results were recorded in the sample receipt register. Only few samples were retested using rapid test at the national laboratory.
Description of fluorescent antibody test
The FAT was conducted at the national veterinary laboratory with little modification of the recommended protocol since positive and negative control were not employed [13]. Briefly, the brain tissue preserved in 50% glycerol saline were washed with buffered saline to remove the glycerol. The brain tissue impression smears were prepared, air dried and were fixed in chilled acetone for 15-30 minutes. The impression smear slides were air dried and 150 µl of fluorescein isothiocyanate (FITC) conjugate anti-rabies antibody (LIGHT DIAGNOSTICS Rabies DFA Reagent, USA) which was prepared by diluting at 1:20 ratio with PBS (PH 7.4) was added on smear and incubated at 370C for 30-45 min in humidified dark chamber. The slides were then washed with PBS thrice for 5 min each which will washed away the unbound antibody and then washed with distilled water to stop the reaction. After washing, slides were air-dried and mounted in 10-20 % glycerine buffer (pH 7.4) or Faramount, Aqueous Mounting Medium, Ready-to-Use from Dako (Dako North America, Inc, Carpinteria, California, USA) and covered with coverslip. The slides were examined under fluorescent microscope at 20X. The presence of apple green fluorescence was considered as positive results (Figure 3). The FAT result was then communicated to the field veterinary officials via e-mail and through telephone which were further communicated to the animal owners or people that had linked with exposure and with the human hospital. If the test is negative, the PEP is normally discontinued.
Data source in this study
Both the rapid test and FAT results status are maintained at the national laboratory register against the unique sample ID. We retrieved the test results data from the database for the period between 2012 and 2017. The data that contain both rapid test and FAT test results status were included for analysis. Any data that contain only FAT results but not rapid test result status or vice versa were excluded for analysis. Therefore, a total of 179 brain tissue samples data recorded between 2012-2017 from different species of animals (113 dogs, 50 cattle, 10 cats, 3 goats, 2 horses, and 1 bear) met the criteria for the analysis (Figure 4, Table 3).
Table 3: Total number of brain tissue samples collected and tested using both rapid test and fluorescence antibody test between 2012 and 2017.
Year
|
Number of samples (percent)
|
2012
|
6 (3.35)
|
2013
|
25 (13.97)
|
2014
|
14 (7.82)
|
2015
|
9 (5.03)
|
2016
|
36 (20.11)
|
2017
|
89 (49.72)
|
Total
|
179 (100)
|
Data analysis
The data management and analysis were done in Microsoft excel 2010 (Redmond Microsoft, USA) and Stata version 14.0 (Stata Corp, USA). A two by two table was constructed to calculate the test sensitivity, specificity, positive predictive value, negative predictive value of the rapid Ag detection test relative to the FAT. The confidence intervals for the sensitivity and specificity were calculated by using the exact binomial distribution.
The overall test agreement between the rapid Ag and FAT was calculated using Kappa test where Kappa measure agreement between the tests. The Kappa value was interpreted as: below 0.0 poor agreement; 0.00 – 0.20 slight agreement; 0.21 – 0.40 fair agreement; 0.41 – 0.60 moderate agreement; 0.61 – 0.80 substantial agreement and 0.81 – 1.00 almost perfect agreement [30]. A pair of agreement measures, positive percent agreement (PPA) and negative percent agreement (NPA) which are analogous to a sensitivity and specificity calculations were also calculated to capture the difference in agreement between samples that are positive and negative according to the reference standard (FAT).