DEVELOPMENT OF A PROTECTIVE INDEX TO RANK EFFECTIVENESS OF MULTIPLE TREATMENTS WITHIN AN EXPERIMENT: APPLICATION TO A CROSS PROTECTION STUDY OF SEVERAL STRAIN OF EIMERIA MAXIMA AND A LIVE VACCINE.

Authors: Allen, Patricia; Danforth, Harry; VINEYARD, B - BIOMETRICS BELTSVILLE MD

Submitted to: Avian Diseases
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/14/2003
Publication Date: 1/12/2004
Citation: Allen, P.C., Danforth, H.D., Vineyard, B.L. 2004. Development of a protective index to rank effectiveness of multiple treatments within an experiment: application to a cross protection study of several strain of eimeria maxima and a live vaccine. Avian Diseases. 48:370-375.

Interpretive Summary: Development of resistance of avian coccidia to prophylactic anticoccidial medication has stimulated a wider use of live oocyst vaccination by poultry producers. However, problems occur because antigenic variation in some Eimeria strains reduces the protective effects of the vaccine. This variation occurs most often in E. maxima where some vaccine strains do not protect effectively against some geographically isolated strains. Vaccine strain candidates need to be experimentally tested for their ability to protect against the widest variety strains before they can incorporated into a highly effective vaccine. Although measurements of challenge-induced weight gain depression and lesion score are used most often to assess vaccine effectiveness, values of other variables such as plasma carotenoids, and NO2-+NO3- can provide important information about effects of challenge infection and the protection provided by the vaccine. In this laboratory, we have found significant correlations among these four independently measured variables. However, in comparing effectiveness of treatments among many experimental groups, it is often difficult to judge which is best based upon four separate measurements. We sought to combine measurements of the above-mentioned variables in such a way as to generate one value that would be a dependent variable more fully reflective the state of each chicken . This variable that we call PX is calculated using the formula : PX = (Ngain+Ncarotenoids) ¿ (Nsqrls+NNO2-+NO3- ). The prefix N indicates that the variables were normalized against the means from an untreated control group. Nsqrls indicates lesion scores that were first transformed to square root values before normalization. Statistical analysis of the PX values using the Proc Mixed procedure of SAS allows comparison of each treatment group against all the others and provides a means to rank the protective ability of vaccine treatments. In this study, mean PX values of unchallenged groups cluster around 0. Mean PX values of protected chickens are statistically close to those from unchallenged groups whereas unprotected chickens have highly negative mean PX values.

Technical Abstract: Vaccination of chickens with live oocysts has become a more widely used method for controlling avian coccidiosis as resistance to anticoccidial medication increases. However, some coccidia strains are not useful in multispecies vaccines because antigenic variation has made them generally less protective. In order to experimentally test a number of strains for the best cross protection, we have devised an evaluation method using four independently measured variables, weight gain, lesion score, plasma carotenoids, and plasma NO2-+NO3-. These values, when measured at 6 days post challenge tend to be significantly correlated. A protective index (PX) is calculated for each chicken using the algorithm : PX = (Ngain + Ncarotenoids)- (Nsqrls+NNO2-+NO3- ) where the prefix N indicates values for a variable normalized against a mean of that variable from a control group. Nsqrls values are normalized values of the square roots of lesion scores. The PX can then be treated as a dependent variable. In this study, mean PX values of unchallenged groups cluster around 0. Mean PX values of protected chickens are statistically close to those from unchallenged groups whereas unprotected chickens have highly negative mean PX values. We are currently using this method of comparing experimental treatments to evaluate efficacies of recombinant vaccine treatments and resistance to anticoccidial drugs.