In view of raised levels of endothelial markers in coronary artery disease (CAD), the aim of the present study was to investigate the status of tissue factor pathway inhibitor (TFPI), another endothelium-associated glycoprotein and coagulation protease inhibitor, in CAD. The intravascular pool of TFPI is heterogeneous with respect to assay-dependent activity. While the standard amidolytic assay works well with both full-length and truncated (lipoprotein-associated) TFPI, the anticoagulant assay works better with the former. The anticoagulant activity of TFPI can be estimated using dilute tissue factor (TF) to trigger clotting of plasma. In the present study, recombinant TF diluted 2,000-fold was used to initiate coagulation. A dose-dependent shortening of clotting time of normal plasma pools with polyclonal antibody against the C-terminal but not the N-terminal peptide of TFPI demonstrated the importance of the C-terminal region, and hence that of full-length TFPI, in conferring its anticoagulant activity, corroborating current opinion. As a further confirmation, the C-terminal peptide itself prolonged dilute TF clotting time of normal pooled plasma in a concentration-dependent manner. The amidolytic and anticoagulant activities of TFPI were determined in 20 patients with clinically and angiographically assessed CAD and in 68 asymptomatic controls. The mean +/- SD ages of patients and controls were 54.9 +/- 10.3 and 48.8 +/- 11.6 years, respectively, the difference being statistically significant (P = 0.04). The mean TFPI activity measured by amidolytic assay was comparable for patients and controls (1.2 +/- 0.3 and 1.3 +/- 0.5 U/ml, respectively). However, the dilute TF clotting time was 115 +/- 26 s in patients, against 99 +/- 10 s in controls (P < 0.0001, irrespective of age adjustment). Since none of the patients had received heparin or had coagulation factor deficiency that may interfere with the assay, prolongation of clotting time may be attributed to the presence of TFPI, particularly the full-length form. To verify this inference, 33 extra aliquots left over from 88 samples (62.5%), 21 from controls and 12 from patients, were incubated with 1:10 diluted antibody against the C-terminal peptide of TFPI prior to dilute TF assay. The mean clotting time of both patients and controls decreased, and the between-group difference leveled (90 +/- 10 versus 88 +/- 20 s for controls and patients, respectively; P = 0.841). The mean drop in clotting time was 9% for the controls and 24% for the patients. This illustrates the specificity of dilute TF assay for full-length TFPI and supports the conclusion that relative to lipoprotein-associated TFPI, the proportion of the full-length form was possibly greater in patients with CAD. Contribution of lipoprotein-associated TFPI to the overall anticoagulant activity by its activated factor X-dependent inhibition of activated factor VII-TF complex seems less important considering the similar between-group mean amidolytic activities.