Multi-test analysis and model-based estimation of the prevalence of Taenia saginata cysticercus infection in naturally infected dairy cows in the absence of a 'gold standard' reference test

R.M. Eichenberger, F. Lewis, S. Gabriël, P. Dorny, P.R. Torgerson, P. Deplazes

Research output: Contribution to journalA1: Web of Science-article


The diagnostic values of seven serological tests (ELISAs) and of the obligatory European Union-approved routine visual meat inspection for the detection of Taenia saginata cysticercosis were investigated. A total of 793 slaughtered dairy cows were selected in three European Union approved abattoirs in Switzerland, an endemic area (apparent prevalence by enhanced meat inspection up to 4.5%) with typically low parasite burdens. ELISAs based on a somatic larval antigen, isoelectric focused somatic larval antigen, larval excretory/secretory antigens, peptide HP6-2, peptide Ts45S-10, pooled peptide solution and a monoclonal antibody antigen capture assay were initially screened. As there is no perfect diagnostic 'gold standard' reference test, the obligatory meat inspection and four selected serological tests were further analysed using Bayesian inference to estimate the "true" prevalence and the diagnostic test sensitivities and specificities. The ELISA for specific antibody detection based on excretory/secretory antigens showed highest sensitivity and specificity with 81.6% (95% credible interval: 70-92) and 96.3% (95% credible interval: 94-99), respectively. The Bayesian model estimated the specificity of the ELISA, based on the synthetic peptide Ts45S-10 as 55.2% (95% credible interval: 46-65) and sensitivity as 84.7% (95% credible interval: 82-88). The sensitivity of the ELISA based on mAbs, detecting circulating antigen, was 14.3% (95% credible interval: 9-23) with a specificity of 93.7% (95% credible interval: 92-96). The diagnostic sensitivity of the obligatory standard European Union meat inspection procedure for the detection of T. saginata cysticercus infection at the abattoir was estimated to be 15.6% (95% credible interval: 10-23). Based on these data, the modelled prevalence of cysticercosis in dairy cows presented at abattoirs in Switzerland was estimated to be 16.5% (95% credible interval: 13-21). These cattle also had a high prevalence of infection with Dicrocoelium dendriticum (60.8%) and Fasciola hepatica (13.5%).
Original languageEnglish
JournalInternational Journal for Parasitology
Issue number10
Pages (from-to)853-859
Number of pages7
Publication statusPublished - 2013


  • Animal diseases
  • Zoonoses
  • Cysticercosis
  • Cysticercus bovis
  • Dicrocoelium dendriticum
  • Fasciola hepatica
  • Cows
  • Epidemiology
  • Prevalence
  • Estimation
  • Serology
  • Visual inspection
  • Meat inspection
  • Evaluation
  • Sensitivity
  • Specificity
  • European Union
  • Standards
  • Switzerland
  • Europe-Central


Dive into the research topics of 'Multi-test analysis and model-based estimation of the prevalence of <i>Taenia saginata cysticercus</i> infection in naturally infected dairy cows in the absence of a 'gold standard' reference test'. Together they form a unique fingerprint.

Cite this