Appendix D - Not part of the original dissertation
Use of Egg Yolk Samples for Predicting age at which to Vaccinate against Infectious Bursal Disease in broiler chickens.
Limited preliminary work reported in the body of the dissertation (Vaccination Date Prediction for Gumboro Disease), suggested that broadly similar results might be obtained as compared to serum samples obtained at day old. For three years it was agreed that up to 40% of the 100 samples targeted from each parent flock could be taken in this way. This report shows a summary of the data generated in this way. Throughout the period in question control of the disease has been exceptionally good.
Materials and Methods: In all 409 groups of samples (276 sera, 133 yolk extracts) were examined. The yolks were examined by diluting a sample of neat yolk obtained with a micropipette from just underneath the yolk membrane as if it were serum. Each group was usually composed of 20 samples. Testing of all samples was in accordance with the kit protocol. Internal laboratory controls were used routinely for QA purposes.
Results: Table 1 summarises the results obtained. There are 2 obvious effects. Yolk samples yielded vaccination date predictions about 5 days higher than equivalent serum samples. Assuming that no more than 40% of samples are from yolk this could result in a maximum increase in average prediction of 2 days - delaying vaccination by this time is unlikely to have a marked effect unless there is heavy environmental contamination with virulent virus. The second obvious effect was that samples tested in 1999 had a VDP about 4 days greater than either of the previous 2 years. Whether this is due to a change in response to vaccination or improved kit sensitivity at high titres is as yet unknown.
| Table 1 Summary of Vaccination date predictions (days) for all samples from one company over a three year period, according to whether the sample was derived from day-old chick serum or unincubated egg yolk. | |||||||||
| Year | Serum Samples | Yolk Samples | Difference | ||||||
| No.Groups | Mean | St.Dev | No.Groups | Mean | St.Dev. | Days | |||
| 97 | 69 | 17.9 | 3.8 | 42 | 22.7 | 3.5 | +4.8 | ||
| 98 | 86 | 17.1 | 4.4 | 49 | 22.8 | 3.8 | +5.7 | ||
| 99 | 121 | 21.5 | 4.3 | 42 | 27.0 | 3.6 | +5.5 | ||
| Overall | 276 | 19.2 | 4.7 | 133 | 24.1 | 4.2 | +4.9 | ||
Regression analysis was also carried out for those groups for which age of parent flock was identified at sample submission. The overall mean predictions were 18.7 and 24.1 days respectively for sera and yolks and there was a slight tendency for both sets of samples to reduce with increasing age. In the following graphs each point represents the predication generated by a single group of samples.
Conclusions: These data suggest that we can expect yolk samples to provide predictions which are about 5 days higher than sera on average. One approach is to simply correct all predictions by this amount, however the difference is likely to be lower for low titre flocks and higher for high titre flocks. A better approach is to titrate a group of yolk extracts by testing duplicate samples at a range of sample dilutions with a view to establishing a revised sample dilution protocol to correct for this effect (I guess, about 1:750, instead of 1:500).