Laboratory tests for anticardiolipin antibodies (aCL) and anti-β2glycoprotein I antibodies (a-β2GPI) face problems common to many autoantibody assays: the lack of a reference standard and the need for each laboratory to assess assay-specific cut-off values. The aims of the study were to evaluate the reference range upper limits (99th percentile) used for aCL and a-β2GPI in the northwest of Italy and to investigate the analytical performances of these assays with the newly obtained reference ranges. We assayed aCL and a-β2GPI in 104 serum samples from patients without a history of thrombosis, pregnancy morbidity, tumours, infections and/or autoimmune diseases (30 males and 74 non-pregnant females). We tested all the commercial assays available in our regions (i.e. Orgentec Diagnostika, Aesku Diagnostics and Inova Diagnostics ELISA; CliA Zenit-RA and EliA Phadia Laboratory Systems). A further 30 serum samples, including 10 from healthy subjects, 10 from antiphospholipid syndrome (APS) patients and 10 from septic patients were assessed to investigate the analytical performance of the obtained cut-off limits. Reference range upper limits obtained with the commercial kits differ among assays and from the values reported by the manufacturer. Moreover, normal reference ranges calculated for IgG and IgM aCL differed from the arbitrary selected laboratory classification values suggested in the guidelines of 40 GPL and MPL.