Specificity of non-structural protein enzyme-linked immunosorbent assays for the detection of serum antibodies against foot-and-mouth disease virus in a target population in New Zealand

Abstract

AIMS: To determine the diagnostic specificities of two commercial screening ELISA, the Bommeli/IDEXX Chekit FMD 3ABC indirect ELISA (ELISA-1) and the Ceditest FMDV NS blocking ELISA (ELISA-2), for foot-and-mouth disease (FMD) in cattle, sheep and pigs in New Zealand, and to compare them with other published studies. To consider the implications for FMD surveillance of using two ELISA in series, a consideration arising from the absence of a gold standard virus neutralisation test (VNT) in New Zealand.

METHODS: Serum samples from non-infected cattle (n=1,015), sheep (n=1,185), and pigs (n=233) from New Zealand were tested in ELISA-1 and ELISA-2 for the detection of serum antibodies against non-structural proteins of the FMD virus. The ELISA were performed according to the manufacturers' instructions.

RESULTS: The diagnostic specificities for ELISA-1 for cattle, sheep and pigs were 99.9%, 99.7% and 99.6%, respectively, and for ELISA-2 were 99.5%, 99.7% and 99.6%, respectively. False-positive reactors in one ELISA were negative in the other ELISA, and vice versa. Using the diagnostic sensitivity data taken from international studies and the diagnostic specificities calculated in this study resulted in overall specificities of 100% in cattle and sheep using serial test interpretation, and 99.2% and 99.0%, respectively, using parallel test interpretation. Diagnostic sensitivities available in the literature varied considerably, and the associated overall serial sensitivity could be as low as 78.7% in cattle and 33.2% in sheep.

CONCLUSIONS: Diagnostic specificities for both ELISA for the target population were comparable with those obtained in livestock populations elsewhere. In surveillance to re-establish New Zealand's freedom from FMD, the use of two ELISA in series would improve the overall specificity in individual animals. However, care would be required to ensure that herd sensitivity was sufficient to detect infection.

KEY WORDS:

• Enzyme-linked immunosorbent assay
• ELISA
• foot-and-mouth disease
• diagnostic specificity
• diagnostic sensitivity
• serial testing
• parallel testing

Acknowledgements

We wish to thank the Ministry of Agriculture and Forestry, Biosecurity New Zealand, for funding this project. We also wish to thank Cedi Diagnostics (now Prionics AG, Switzerland) and Bommeli Diagnostics (now IDEXX Laboratories, Maine) for supplying us with trial ELISA kits.