Poor anticoagulant response to APC is conveniently screened by a commercially available functional test (Coatest APC Resistance) allowing identification of APC-resistant patients. These patients may then be genotyped with respect to factor V, the Arg -> Gln mutation being the principle cause of APC resistance. However, determination of phenotype generally precedes that of genotype, and the need for an "abnormality threshold" prompted a study of inter-batch variations and the clinical conditions associated with an altered APC response. The response to APC was assessed twice in plasma from 111 patients using two of four successive kit batches. A modest but significant inter-batch variability was observed. At the same time, we also tested 130 patients with retinal venous occlusion (RVO), 28 patients with glaucoma and 24 normal volunteers. The APCaPTT/aPTT ratio was found to be lower in the presence of elevated thrombin-antithrombin complexes (r = 0.167, p < 0.02) and low blood viscosity (at high shear rate: r = 0.305, p < 0.0001) independently of any alteration in genotype.