Search

A study was performed to investigate the genotypes and sub-groups of pestiviruses present in ruminants in Ireland. These comprised one ovine and eighteen bovine pestiviruses from Northern Ireland and six bovine pestiviruses from the Republic of Ireland. A 288 base pair (bp) portion of the 5'-non coding region (5'-NCR) from each of 25 pestiviruses collected over a period of 31 years was amplified by reverse-transcription-polymerase chain reaction (RT-PCR) and the product directly sequenced. From each pestivirus, nucleotide sequences corresponding to bases 130 to 374 of the 5'-NCR of NADL were aligned and compared with each other and with the corresponding sequences of a number of reference, field or vaccinal strains of BVDV types I and II, border disease virus and classical swine fever virus. All of the 25 sequenced pestiviruses were found to be BVDV type Ia. These were closely related to the constituent viruses of the 2 inactivated vaccines currently licensed for use in Northern Ireland and to recent bovine isolates from England.

Bulk tank milk samples collected from 929 Northern Ireland dairy herds were classified according to their geographical location, somatic cell count and milk yield. Each sample was tested by ELISA for antibody levels to bovine viral diarrhoea virus (BVDV) and the herds were assigned to four groups with increasing antibody levels. In nine herds (1 per cent) no antibodies were detectable in the milk samples (group 1) and they were detectable at low levels in the milk from 90 herds (9.7 per cent, group 2), at moderate levels in 369 herds (39.7 per cent, group 3), and at high levels in the remaining 461 herds (49.6 per cent, group 4). The testing of samples from 90 herds in groups 1 and 2 after an interval of 12 months showed that the annual incidence risk for new infections with BVDV was in the range 0.133 to 0.477. There were significant relationships (P < 0.001) between the mean corrected optical density in the ELISA and herd location and somatic cell count, but not yield; the relationship with somatic cell count was linear (P < 0.01). Forty-three group 3 herds and 49 group 4 herds were selected at random and bulk milk samples were tested for BVDV by reverse transcriptase polymerase chain reaction. None of the group 3 herds was positive, but five group 4 herds were positive.

The suitability of a commercial bovine viral diarrhoea virus (BVDV) antigen capture enzyme-linked immunosorbent assay (ELISA) for routine diagnostic testing of bovine serum samples was evaluated by comparing the ELISA results of 214 sera with those obtained after two passages in roller tube cultures of fetal bovine lung cells and immunofluorescent staining using fluorescein isothiocyanate-conjugated hyperimmune BVDV anti-serum. In addition, 208 of these samples were tested by virus isolation in a microtitre system followed by indirect immunoperoxidase staining using a pool of two non-competing pestivirus specific monoclonal antibodies. The sensitivity of the ELISA compared with virus isolation followed by immunofluorescent and immunoperoxidase staining was 47.8 and 45.8%, respectively. The corresponding figures of specificity and overall correlation were 95.3 and 95.1%, and 90.2 and 89.4%. Twenty-two of 24 pestivirus isolates from the positive blood samples were typed as BVDV-like by monoclonal antibodies, indicating that the poor sensitivity of the ELISA was not due to the presence of atypical pestiviruses in the test sample. These results suggest that this ELISA is not suitable for testing blood samples for BVDV in a diagnostic laboratory.