Abstract Background Brucellosis is frequently reported among wild boar populations in Europe. The aim of the study was to assess the epidemiological situation in Belgium, regarding the steady increase of wild boar populations over the last decades. Several serological tests were used and compared with culture and IS 711 polymerase chain reaction (PCR), to determine the most suitable combination of diagnostic tools for conducting a successful prevalence study in wildlife. Results An indirect enzyme-linked immunosorbent assay (iELISA) was used on 1168 sera from hunter-killed wild boar sampled between 2003 and 2007 in 4 natural regions of southern Belgium. Results gave an apparent prevalence of 54.88% (95% CI 52.03-57.73). Prevalence was significantly affected by age and by the year of study, but not by sex nor by the region of sampling. The relative sensitivities of the complement fixation test (CFT), the Rose Bengal test (RBT), and the slow agglutination test (SAT) versus the iELISA differed widely between tests, reaching 62.67%, 46.68%, and 34.77%, respectively. The relative specificities of the CFT, RBT and SAT versus the iELISA were respectively 99.01%, 92.49%, and 99.1%. From seropositive animals (iELISA), 9% were positive by culture and 24% by PCR when testing spleen and/or tonsils. Sensitivity of the PCR was higher on tonsils than on spleen. All bacterial isolates were identified as Brucella suis biovar 2. Conclusions Brucellosis is widespread among wild boar in southern Belgium, with seroprevalences having increased over ten years, and constitutes a growing risk of spillback to outdoor-farmed pig herds. The iELISA showed a better sensitivity than the CFT, RBT and SAT. Serological tests must be associated with direct diagnosis and PCR proved more sensitive than culture under wildlife sampling conditions. Spleen and tonsils are lymphoid tissues usually sampled in multi-disease monitoring programs. They remain top-grade organs for direct diagnosis of brucellosis, with a preference for tonsils.