by Paul Loubet, Chloé Magnan, Florian Salipante, Théo Pastre, Anne Keriel, David O’Callaghan, Albert Sotto, Jean-Philippe Lavigne
IntroductionBrucellosis, a zoonotic infectious disease caused by bacteria of the genus Brucella, remains a significant global health concern in many parts of the world. Traditional diagnostic methods, including serological tests, suffer from limitations, including low sensibility and high false-positive rates, emphasizing the need for improved diagnostic strategies. In this study, we aimed to optimize diagnostic accuracy by reevaluating serological tests and exploring novel diagnostic algorithms.
MethodsA retrospective observational study was conducted using sera collected between June 2012 and June 2023 at the French National Reference Center for Brucella. Various serological tests, including Rose Bengal plate test (RBT), standard agglutination test (SAT), Brucellacapt, and ELISA for IgM and IgG, were performed. Different diagnostic algorithms were evaluated, combining RBT with SAT, Brucellacapt, and ELISA to enhance the performance of diagnostic tests.
ResultsAmong 3587 sera analyzed, 148 were confirmed cases of human brucellosis. Individual serological tests exhibited good sensitivity and specificity but lacked diagnostic accuracy. However, combining RBT with SAT or Brucellacapt significantly improved diagnostic performance, with reduced false positives. The most promising results were observed when an algorithm was built combining RBT, Brucellacapt, and ELISA for IgM and IgG (a score value of 0.5 with 90.5% for sensitivity, 99.7% for specificity, 92.4% for PPV, and 99.6% for NPV).
ConclusionsSerological tests remain crucial for brucellosis diagnosis, but their limitations necessitate innovative diagnostic approaches. Combining multiple serological tests in diagnostic algorithms shows promise in improving diagnostic accuracy. Efforts to refine diagnostic, strengthen surveillance, and raise awareness are essential for effective brucellosis control, particularly in resource-limited settings.
