Vaccination Date Prediction for Gumboro Disease

Elisa assay of maternal immunity in young chicks is being increasingly used as an aid in predicting the date at which the chicks will become sufficiently susceptible to enable efficient vaccination. The concept was first investigated in White Leghorns by Solano and others (1986). A formula for predicting day to vaccinate was worked out and validated in large-scale broiler trials by scientists at the Doorn Institute in the Netherlands (Kouwenhoven, 1991).

The formula is derived from the regression of expected reduction in maternal antibody levels. In one of its simplest forms the formula for broilers can be given as :

Square Root(Measured Mean Titre)- Square Root(Target titre)
---------------------------------------------------------------- = VDP
============2.82

The divisor is a constant derived from antibody half-life. A modification of the formula would be required for use with broiler parents and layers to take into account differing growth rates and rates of depletion of maternal antibody. We suggest adding 10% for broiler parents and 20% for layers, though it must be emphasized that this technique has not been validated in these classes of chicken.

The VDP value given by the formula is the interval in days between the day of sampling and the day on which it can be expected that the mean titre will reach the designated target titre.

Differing target titres can be chosen depending on the degree to which the vaccine in use is affected by maternal antibody. The target titres which have been used in the UK are :

As the VDP value is the interval in days from sampling to earliest day to vaccinate then the age of the birds on the day of sampling must be added to get the age from which vaccination may be used. For example, if the chicks are sampled at 1 day of age and the VDP is 17 then the flock could be vaccinated from 18 days of age.

The prediction attempts to time vaccination to 50% susceptibility. If the titres are highly variable we may advise a modified schedule rather than following the prediction strictly. On the other hand, if early challenge is not expected then every consideration should be given to delaying vaccination beyond the day predicted in order to further improve the vaccine "take".

We have been carrying out assays for IBD antibody levels in newly-hatched broiler and layer chicks for over 2 years. Our impression is that control of the disease is enhanced when the predictions are taken into account. To date it has been customary to repeat sample the progeny of each parent flock at intervals through the period of production (usually 20 chicks per month). This was reasonable since it is known that the level of antibodies circulating in the hens blood decreases with age (p.33), and that the level of antibodies present in yolk is roughly proportional to that in the blood (p.15).

This system presents some problems :

1. Predictions fluctuate up and down with each successive test. Many of these fluctuations seem to be random (they may be related to the maturity of the chicks sampled, sample handling, assay kit batch etc.). For this reason we have advised that the "trend" be used rather than to rely on a particular report.

2. Because the information on a flock is generated and distributed over a long period there can be confusion and delays in getting the information to the people who have to act on it (the farm managers).

3. Blood sampling of a substantial number of chicks by hatchery staff by decapitation is required.

As a first step in deciding how to resolve these problems we have carried out an analysis of vaccination date predictions according to parent flock age. One broiler company (A) was chosen for its traditionally high levels of antibody and another (B) because its chicks tend to have moderate to good levels of maternal antibody .

One hundred and forty eight groups of chicks from company A were included in the study (average prediction 19.3 days), and 276 groups from Company B (average prediction 16.2 days). Figures 3.26 and 3.27 are scatter graphs for each company and include a line showing the regression of prediction on age of the parent flock. There is a slight downward trend in the chicks from company A, and an even slighter downward trend in the chicks from company B. This demonstrates that progeny antibody levels do not fall off dramatically as the parent flock ages. It has been our impression for some time that, if we exclude the "random" variation, then it is possible to establish a prediction for a flock early in production and use it throughout the normal life of a flock. These data suggest that intensive sampling at the beginning of production can determine a prediction which will not change substantially with age. For very high predictions (over 18-20 days) there is the option of cutting a day after about 15 weeks production.}

In order to determine if the effect is the same in layer chicks we have extracted the more limited data from one layer hatchery. In this case only 22 groups of chicks were sampled and the mean prediction was 17.3 days. These data are summarized in Figure 3.28 below. There is no evidence of reducing prediction as the parent flocks age in this case. The minimal reduction in maternal antibody transfer with increasing age of parent flock was unexpected. The pattern of reducing circulating antibody levels in parent blood (see page 33) should lead to a reduction of VDP of the order of 7 days over the period of egg production. The explanation for this effect probably lies in the fact that egg yolk weight, as a proportion of chick weight, increases as the parents age (Marion et.al. 1966). The efficiency of deposition of antibodies and other proteins in yolk also increases as the flock matures (Kramer and Cho,1970).}

In the hope that we may be able to reduce the need to sacrifice day-old chicks for blood sampling we have carried out some comparisons between predictions based on fresh yolk and those based on day-old bloods. Figure 3.29 shows the difference between the yolk prediction and the mean blood predictions for the same flock, according to the age at which the eggs were collected. Testing of egg yolk seems to provide as reliable an indicator of VDP as testing day-old chick blood. The values obtained with yolk tests seem to correlate most closely with those from blood tests early in production.

Analysis of the data presented here led to the development of a revised sampling procedure for VDP. In summary this involves the sampling of each new flock on at least 5 occasions up to 30 weeks of age. Up to 40% of samples could be fresh eggs and at least 20 samples should be submitted on each occasion. However see Appendix D - VDP and Yolk samples (not part or original dissertation).

New flocks coming into production should be assigned the historical average figure for vaccination date predictions for the supply flock area until such time as the actual predictions become available. As soon as 2 prediction results become available an "MGV" value should be calculated. The MGV (Minimum Gumboro Vaccination) value reported by the laboratory for each group will be the VDP + the age of birds when sampled (egg yolks to be considered 1 day of age). It should be updated with each new result up to a maximum of 5. The MGV value is calculated as follows :

MGV = (Arithmetic Mean of all available MGV values for flock)

Ideally the calculation and reporting components of the system should be included in any hatchery data-management system. The implementation or continued use of this system is likely to be of most benefit in areas suffering losses from Acute Gumboro disease. Ideally the MGV (Minimum Gumboro Vaccine) value should be informed along with the flock code to the growing farm manager. This should be regarded as the earliest age at which the broiler flock can be effectively vaccinated.

Our attention has been focused here on the practicalities of the use of a formula for predicting loss of maternal anti-IBD antibody in order to help achieve a better response from vaccination. There may well be other applications for the same concept. The formula could be used for other diseases in which maternal antibody is known to affect vaccine take. It can also be used as an aid in interpreting serological data. If, for example, positive titres for any disease are observed at 21 days of age, they may be due to maternal antibody or to an active response to challenge or vaccination. If the titres of maternal antibody present at 1 day old are known then the formula may be used to predict whether the 21 day titres are of maternal origin.