Comparison of 2 ELISAs for detecting exposure to Brucella ovis. (September 2020)

• Record Type: Journal Article
• Authors: Elderbrook, Molly J.; Schumaker, Brant A.; Ueti, Massaro W.; Bastos de Almeida, Meila; Vieira, Thallitha S. W. J.; Vieira, Rafael F. C.; Sondgeroth, Kerry S.
• Is Part Of: Journal of veterinary diagnostic investigation. Volume 32:Number 5(2020:Sep.)
• Page Start: 700
• Page End: 705
• Publication Date: 2020-09
• Subjects: agar gel immunodiffusion -- Brucella ovis -- ELISA -- ram epididymitis -- sheep; Veterinary medicine -- Diagnosis -- Periodicals; 636.0896075
• DOI: 10.1177/1040638720943880
• Languages: English
• ISSNs: 1040-6387

Abstract

Control of Brucella ovis infection in sheep flocks in the United States depends on early detection of B. ovis antibodies via serologic testing. We used 2, 276 sheep sera and various cutoff values to compare seroprevalence and agreement between 2 ELISAs: the National Veterinary Services Laboratories (NVSL) B. ovis indirect ELISA and the IDEXX B. ovis ELISA kit. A subset of 295 sera was used to compare agreement and evaluate relative sensitivity and specificity of the 2 ELISAs with an agar gel immunodiffusion (AGID) test kit. There was no significant difference in B. ovis seroprevalence between the ELISAs; however, there was poor agreement between them. When the AGID test was used as the reference test, the IDEXX ELISA with a moderate cutoff value (S/P ratio = 45%) had the highest relative sensitivity of 38.1% and specificity of 92.0%. The NVSL ELISA with a lax cutoff value (S/P ratio = 0.75) had relative sensitivity of 19.1% and specificity of 94.6%. Receiver operating characteristic analysis revealed that optimal cutoff values for the NVSL and IDEXX ELISAs were 0.091 and 16.5%, respectively. This results in sensitivity and specificity of 85.7% and 31.8% for the NVSL ELISA, and sensitivity and specificity of 81.0% and 53.6% for the IDEXX ELISA, respectively.