A retrospective evaluation of pooled serum ELISA testing in the frame of the French eradication program for infectious bovine rhinotracheitis.

Pooled serum testing using whole-virus indirect ELISA has been recently recognized as an official method for surveillance of bovine herpesvirus 1 (BoHV1) in cattle herds in Europe. In this study, a retrospective analysis of data from the French BoHV1 surveillance campaign 2018-2019, including 7434 BoHV1-free certified herds and 157 infected herds, was performed in order to evaluate the diagnostic specificity and sensitivity of two pooled serum indirect ELISAs (from IDEXX and IDVet), in comparison with individual testing by blocking ELISAs targeting the gB and gE proteins. Pooled serum testing showed a relative specificity higher than 97.5% and a detection rate of 100% since all gB+/gE+ samples were found in positive pools. At the herd level, no more than one false positive pool was observed in most of BoHV1-free certified herds, leading to a herd relative specificity of 85.1% and 86.0% for the IDEXX and IDVet pooled serum ELISAs, respectively. Among infected herds tested by pool sizes up to 10 sera (n = 122), 46% of herds were detected through pools of size 10 containing a single positive sample, 23% through pools of size 10 containing at least two positive samples, and 31% through pools of smaller sizes. A complementary study based on manually constituted pools revealed that at least one positive sample in 100% and 93.4% of herds could be detected individually by pools of size 10 with the IDEXX and IDVet ELISAs, respectively. However, pooled serum ELISAs were influenced by the level of individual reactivity, since pools composed of either one weak-positive sample or one gB+/gE- sample could yield negative results. Altogether, these results provided the first evidence that pooled serum testing (pool size up to 10) is a suitable strategy for surveillance of BoHV1-free cattle farms.