Summary: Equine recurrent uveitis (ERU) is an ocular disease characterized mainly by recurrent episodes of inflammation, alternating with quiescent episodes of various durations. There are numerous aetiological theories for the cause of ERU, but to date, the "classic ERU" is most commonly associated with an intraocular leptospiral infection. The pars plana vitrectomy has, so far, been the most promising treatment of ERU and is thought to be more successful in eyes with detectable leptospiral antibodies in the microscopic agglutination test (MAT) and therefore a preselection is recommended. The objective of this study was to investigate whether different laboratories have an influence on the frequency of detection of leptospiral DNA and leptospiral antibodies from vitreal samples of horses with ERU and if these results are in accordance with cultural testing of the vitreal samples. Vitreal samples were obtained from 93 eyes of 78 horses of various breeds and age with a clinical history of ERU. The vitreal samples were taken at the beginning of the vitrectomy to keep the dilution of the vitreal samples as low as possible. The samples were split and sent to three different laboratories in Germany (L1, L2 and L3). Laboratory 1 (L1) performed a polymerase chain reaction (PCR) and a MAT, laboratory 2 (L2) carried out a PCR and laboratory 3 (L3) performed a MAT, a cultural examination and an enzyme-linked immunosorbent assay on the vitreal specimens. The amount of dilution was evaluated by comparing the urea content of 34 vitreal samples with the urea content of the serum of each of the horses at the time of sampling. Furthermore, a serial dilution was performed with undiluted vitreal material of two enucleated eyes and sent to L1 for MAT and PCR. The results of the PCR indicated a strong conformity between L1 and L2 (Kappa 0.95), both laboratories concurrently revealed 52 positive and 31 negative results and only two specimens showed deviating results. The MAT in L1 was positive in 51% of the vitreal samples tested, whereas only 41% of the vitreal samples were tested positive in L3. The conformity was, moreover, serovar-and breed-dependent. The Kappa was 0.67 when only including serovar Grippotyphosa, and was 0.82 for serovar Pomona (Tab. 5). Warmblood horses displayed a Kappa of 0.56 between L1 and L3, whereas the Kappa for the Icelandic horses was 0.91 between L1 and L3. The cultivation of leptospires was successful in 16 % of the vitreal specimens. The dilution of vitreal samples was measured to range between 1 and 2.1 with a standard deviation of 1.68. Overall, 67 vitreal samples were tested simultaneously in all three laboratories. A positive result in at least one of the tests performed occurred in 57 % (38), whereas 43 % (29) of the vitreal specimens were tested negative in all five tests performed (PCR in L1 and L2, MAT in L1 and L3, and culture in L3). The variation of agreement of the MAT between serovar Grippotyphosa and serovar Pomona and between Warmblood and Icelandic horses in L1 and L3 might be account...