Challenges for bovine viral diarrhoea virus antibody detection in bulk milk by antibody enzyme-linked immunosorbent assays due to changes in milk production levels

Abstract

Background: Bovine viral diarrhoea (BVD) is considered eradicated from Denmark. Currently, very few (if any) Danish cattle herds could be infected with BVD virus (BVDV). The Danish antibody blocking enzyme-linked immunosorbent assay (ELISA) has been successfully used during the Danish BVD eradication program, initiated in 1994. During the last decade, the cattle herd size has increased while the prevalence of BVDV has decreased. In this study, we investigated how these changes could affect the performance of the Danish blocking ELISA and of the SVANOVIR® BVDV-Ab indirect ELISA. The latter has successfully been used to eradicate BVD in Sweden. Data (2003-2010) on changes in median herd size and milk production levels, occurrence of viremic animals and bulk milk surveillance were analysed. Additionally, the Danish blocking ELISA and the SVANOVIR ELISA were compared analyzing milk and serum samples. The prevalence of antibody positive milking cows that could be detected by each test was estimated, by diluting positive individual milk samples and making artificial milk pools.

Results: During the study period, the median herd size increased from 74 (2003) to 127 cows (2010), while the prevalence of BVDV infected herds decreased from 0.51 to 0.02 %. The daily milk yield contribution of a single seropositive cow to the entire daily bulk milk was reduced from 1.61 % in 2003 to 0.95 % in 2010 due to the increased herd size. It was observed that antibody levels in bulk milk decreased at national level. Moreover, we found that when testing bulk milk, the SVANOVIR® BVDV-Ab can detect a lower prevalence of seropositive lactating cows, compared to the Danish blocking ELISA (0.78 % vs. 50 %). Values in the SVANOVIR® BVDV-Ab better relate to low concentrations of antibody positive milk (R(2) = 94-98 %), than values in the blocking ELISA (R(2) = 23-75 %). For sera, the two ELISAs performed equally well.

Conclusions: The SVANOVIR ELISA is recommended for analysis of bulk milk samples in the current Danish situation, since infected dairy herds e.g. due to import of infected cattle can be detected shortly after BVDV introduction, when only few lactating cows have seroconverted. In sera, the two ELISAs can be used interchangeably.

Fig. 1

Changes in the size and milk production of Danish dairy herds from 2003 to 2010. Cows/Herd = median herd size (divided by 10); Milk/Herd = milk produced per herd (in 100,000 kg). Herds/Year = number of Danish dairy herds (in 1,000), which delivered milk from January to December; Milk/Year = national milk production (in billion kg of milk)

Fig. 2

Results obtained on diluted individual milk samples. On the x-axis, 1 corresponds to the undiluted sample, while 2-8 represent dilution steps 1/2 up to 1/128. □ = mean bl% using the Danish blocking ELISA; ∆ = mean PP-value using the SVANOVIR. Grey bars represent 95 % confidence interval around each mean, H: highly positive group (n = 4), M: medium positive group (n = 3), and L: low positive group (n = 3). Horizontal dashed lines represent the cut-offs (bl% = 0 and PP = 2), above which samples were classified as antibody positive. N.B. Test values below zero do not have any biological meaning, and thus, in the 95 % confidence intervals we set the minimum bar limit to 0. We proceeded in a similar way for the maximum limit of the blocking ELISA, which cannot have values > 100 %

Fig. 3

Results obtained on diluted individual serum samples. On the x-axis, 1 corresponds to the undiluted sample, while 2-8 represent dilution steps 1/2 up to 1/128. □ = mean bl% with the Danish blocking ELISA; ∆ = mean PP-value using the SVANOVIR. Grey bars represent 95 % confidence interval around each mean, H: highly positive group (n = 4), M: medium positive group (n = 3), and L: low positive group (n = 3). Horizontal dashed lines represent the cut-offs (bl% = 50 and PP = 15) above which samples were classified as antibody positive. N.B. Test values below zero do not have any biological meaning, and thus, in the 95 % confidence intervals we set the minimum bar limit to 0. We proceeded in a similar way for the maximum limit of the blocking ELISA, which cannot have values > 100 %

Fig. 4

Change in bulk milk values after removal of all PI calves (herds B and C). Results from two dairy herds during a 149 days evaluation for herd B and 915 days for herd C. The y-axis represents the bl% and PP-values according to the blocking ELISA and the SVANOVIR respectively, while the x-axis represents the number of days since removal of the last born PI calf and the bulk milk sampling. Four and fifteen samples were tested from herds B and C respectively