Resistance to activated protein C (APC) is due, in most cases, to a G to A mutation at nucleotide 1691 of factor V (FV) gene (the Leiden mutation). This inherited abnormality is now considered to be the major hereditary cause associated with an elevated risk of thrombosis. For this reason, laboratories are faced with an increasing number of samples referred for APC resistance diagnosis. This could have serious economic consequences and a comprehensive laboratory screening strategy for APC resistance is necessary. An original DNA assay based on denaturing gradient gel electrophoresis (DGGE) was designed in our laboratory. During a first period we systematically performed DNA analysis and compared the results with phenotypic assays. Using the modified functional test with a 1:5 predilution of plasmas, the cut-off value for APC resistance ratio was 2.6 in our sample. Among 94 consecutive patients referred to our laboratory we found a clear cut-off between the APC resistance ratio obtained for normal and abnormal individuals. The modified test had a predictive value of 1.0 found by a cut-off < or = 2.6 for the heterozygote FV Leiden. This obviates the necessity of genotyping subjects with a normal phenotype. Among patients with an abnormal phenotype we were able to fully discriminate between homozygous and heterozygous patients using a cut-off value of 1.5. Nevertheless, our results demonstrate that, because of false-positive results such as lupus anticoagulant, genotyping is still indicated for patients with an abnormal ratio determined with the modified APC resistance test. The strategy described here allows us to safely lower the number of samples analysed by DGGE.