Original Paper Haemostasis 1997;27:163-167 Extrinsic Coagulation Factors and Tissue Factor Pathway Inhibitor in End-Stage Chronic Renal Failure Takefumi Matsuo Masanobu Koide Kazuomi Kario Shunji Suzuki Miyako Matsuo Key Words Abstract End-stage chronic renal failure Factor Xlla Extrinsic coagulation factors Tissue factor pathway inhibitor Diabetes mellitus The relationship between extrinsic coagulation factors, tissue factor pathway inhibitor (TFPI) and activated factor XII (FXIIa) was examined in 71 patients with end-stage chronic renal failure. They had chronic stable uremia due to regular hemodialysis. The patients were divided into two age- and sexmatched groups with and without diabetes mellitus. As extrin sic coagulation parameters, FVIIa and FVII antigen (FVIIag), tissue factor antigen and TFPI (the activity and antigen) were measured. FXIIa was measured as a marker of contact activa tion, and thrombin generation was evaluated using the two markers thrombin-antithrombin III complex and prothrom bin fragment 1 + 2. In both hemodialysis groups with and without diabetes, significant elevations of FXIIa, FVIIa and tissue factor, with high levels of TFPI, were found. Thus, hyperactivation of the coagulation system was in part compen sated by TFPI, and a significant increase in FXIIa could not directly affect FVIIa hyperactivation. No differences of these parameters, except for FVIIag and fragment 1 + 2, were found between the groups with and without diabetes. It is suggested that the long-term hemodialysis might have masked any dif ferences due to the underlying disease in these two subgroups. Introduction In patients undergoing regular hemodialy sis for end-stage renal failure (ESRF), the risk of cardiovascular complication is increased KARGER © 1997 S. Karger AG, Basel 0301 -0147/97/0274-0163S15.00/0 Fax + 41 61 306 12 34 E-Mail email@example.com www.karger.com This article is also accessible online at: http://BioMedNet.com/kargcr with the progression of atherosclerosis. Blood lipid and coagulation abnormalities are major contributing factors in the progression of ath erosclerotic vascular change. In our previous study , we found that ESRF is associated Dr. Takefumi Matsuo Department of Internal Medicine Hyogo Prefectural Awaji Hospital, Shimogamo Sumoto, 656 (Japan) Fax +81 799 24 5704 Downloaded by: Vanderbilt University Library 126.96.36.199 - 10/27/2017 2:04:41 PM Hyogo Prefectural Awaji Hospital, Sumoto, Japan Subjects and Methods Seventy-one consecutive ESRF patients aged 61.4 ± 14.1 (mean ± SD) years, 35 males and 36 females, who had been undergoing regular hemodialysis treat ment at the Awaji Hospital Kidney Center, Sumoto, Japan, for at least 1 year were enrolled in this study. Before hemodialysis, all subjects had uremic syn drome, hyperkalemia, metabolic acidosis, fluid over load and < 10 ml/min creatinine clearance. The coagu lation parameters in the 25 patients with diabetic nephropathy (age 63.5 ± 11.5 years, 12 males and 13 females) were compared with those in 25 sex- and agematched nondiabetic patients (age 63.8 ± 12.1 years; 23 patients with chronic glomerulonephritis and 2 with nephrosclerosis). Thirteen diabetic patients had been diagnosed as having non-insulin-dependent and 12 insulin-dependent diabetes. All diabetic patients were under the care of a diabetologist and had maintained hemoglobin A|C levels under 8.5% (6.4±1.0%). Eighty-six healthy controls (69.4 ± 17.8 years, 39 males and 47 females) who had shown no medical problems at a medical check-up were also studied. Blood samples were obtained in 0.1 vol of 3.8% tri sodium citrate in the disposable siliconized glass tubes 164 Haemostasis 1997;27:163-167 in the morning before the start of hemodialysis. Plas ma was immediately separated and stored at -30° C in plastic tubes until each measurement. Measurement of FXIIa was carried out using a direct enzyme immu noassay (Shield Diagnostics), employing antibody which recognizes a-XIIa and P-XIIa. The intraindivid ual variation of the FXIIa assay in 31 patients with ESRF was assessed using blood samples obtained at 6-month intervals. No significant intraindividual dif ferences were observed, with a positive correlation for the two measurements (r = 0.867; p < 0.001). Tissue factor antigen was measured by an enzyme immunoas say consisting of two monoclonal antibodies against human tissue factor, one side of which was labeled with horseradish peroxidase, biotinyl-tyramide to am plify the enzyme signal and 2,2'-azino-di-(3-ethylbenzthiazoline sulfonate) as a substrate , Plasma FVIIa was directly measured using a previously described fluorogenic assay  employing a fluorogcnic peptide substrate for thrombin (Peptide Institute), congenital human FVII-deficient plasma (George King Bio-Medi cal) and recombinant human soluble tissue factor ex pressed in yeast and purified. The reaction was started by addition of soluble tissue factor. Fluorescence in tensity was transformed into reaction time, which was defined as the time to reach an estimated fluorescence intensity. The reaction time was converted to FVIIa (ng/ml) by comparison with a standard curve produced by serial dilution of purified FVIIa (Chemo-Sero-Therapcutic Institute). FVII antigen (FVIIag) was assayed using a commercially available enzyme-linked immu nosorbent assay (ELISA) kit (Diagnostics Stago). A modified method of that described by Sandset et al.  was employed for tissue factor pathway inhibitor (TPFI) activity (TFPla) determination. TPFI antigen (TFPIag) detected as full-length TFPI was measured using an ELISA kit (American Diagnostics). Commer cially available ELISA kits were employed to measure the thrombin-antithrombin III complex (TAT) and prothrombin activation fragment 1 + 2. Data are showm as means ± SD. Statistical signifi cance was determined by the unpaired Student’s t test. Pearson’s correlation coefficients were calculated for the different variables. Results In the patients with ESRF, FXIIa was sig nificantly increased and did not correlate with the levels of blood urea nitrogen or creatinine. Matsuo/Koide/Kario/Suzuki/Matsuo Downloaded by: Vanderbilt University Library 188.8.131.52 - 10/27/2017 2:04:41 PM with hyperactivation of coagulation due to increased factor VII (FVII) activity and anti gen level and that these abnormalities con tribute to the pathogenesis of cardiovascular complications. The present study was carried out to clarify in more detail the role of factors in the extrinsic coagulation pathway, includ ing activated FXII (FXIIa). The blood of these patients with ESRF is regularly exposed to a foreign surface during hemodialysis un der anticoagulation. FXII undergoes contact activation due to the extracorporeal circuit, being converted to FXIIa , FXIIa directly induces activation of FVII to FVIIa, a useful marker of activation in the extrinsic pathway . Since diabetes mellitus is more frequently associated with severe cardiovascular compli cations, we also examined whether the diabet ic patients showed hyperactivation in the ex trinsic pathway compared to nondiabetic pa tients. Table 1. Contact activation, extrinsic pathway activation and inhibition in patients with ESRF A Extrinsic factors ESRF Controls FXIIa, ng/ml Tissue factor, pg/ml FVIIa, ng/ml FVIIag, % 8.6± 3.7** (n = 59) 2.0± 1.2 (n = 86) 387 ± 110** (n = 50) 130 ±41 (n = 79) 6.3 ± 3.1 ** (n = 69) 3.9 ± 1.4 (n = 86) 135 ±43 (n = 70) 139 ± 37 (n = 81 ) TFPIa, % TFPIag, ng/ml 185 ±37** (n = 71) 99 ± 28 (n = 71) 289 ±22** (n = 69) 120 ±21 (n = 81) TAT, ng/ml Fragment 1+2, nmol/1 5.5 ±6.3* (n = 70) 2.7 ± 3.3 (n = 86) 2.3 ±0.8** (n = 50) 1.0 ±0.2 (n = 50) B TFP1 ESRF Controls C Thrombin generation ESRF Controls *p<0.01, ** p < 0.001 : compared to controls. Tissue factor and FVIIa were also significant ly increased compared to those in the control group, but FVIIag showed no significant in crease (table 1). TFPIa, TPFIag, TAT and fragment 1 + 2 showed significant elevations compared to the control. Although FXIIa was markedly increased in ESRF, it showed no significant correlation with either tissue factor (r = 0.281; p < 0.1) or FVIIa (r = 0.268; p < 0.1); prothrombin activation was observed despite the increase in TFP1. Therefore, there was no significant correlation between each extrinsic factor and thrombin generation markers, and there was no evidence that the increased FXIIa was directly related to the increases in plasma tissue factor and FVIIa levels. ESRF patients were divided into two groups according to whether they had un derlying diabetes or not. No differences were recognized between the patients with and without diabetes, except for TAT (table 2). Significant increases in extrinsic coagulation and inhibitory factors were observed in both subgroups of ESRF patients compared to the control group, except for FVIIag. The results indicated that the marked hyperactivation in the diabetic ESRF group was not more marked than that in the nondiabetic group. Extrinsic Coagulation Factors, FXIIa and ESRF Haemostasis 1997;27:163-167 In this study, FXIIa was found to be signif icantly increased in patients with ESRF. As no correlations were found among extrinsic coagulation factors, FXIIa and inhibitory fac tors, hyperactivation in the extrinsic pathway is thought to occur independently of the in crease in FXIIa. In this regard, Kario et al.  165 Downloaded by: Vanderbilt University Library 184.108.40.206 - 10/27/2017 2:04:41 PM Discussion Table 2. Hemostatic factors in patients with ESRF with and without diabetes mellitus A Extrinsic factors Diabetic ESRF Nondiabetic ESRF Age-matched control FXIIa, ng/ml Tissue factor, pg/ml FVIIa, ng/ml FVIIag, % 9.1 ±3.6° (n = 22) 7.6±2.4C(n = 23) 1.7 ± 1.1 (n = 25) 368 ± 105c (n = 20) 387 ± 116c (n = 19) 128 ±41 (n = 25) 5.8 ± 2.8b (n = 25) 6.6±2.6c (n = 24) 3.5 ± 1.3 (n = 25) 129 ± 27 (n = 24) 140±43 (n = 25) 143 ± 38 (n = 25) TFPIa, % TFPIag, ng/ml 232 ± 32c (n = 20) 217±38c (n = 24) 99 ±29 (n = 20) 266±48c (n = 25) 272±44c (n = 24) 121 ±22 (n = 21) TAT, ng/ml Fragment 1+2, nmol/1 3.3 ± 1.4ad (n = 24) 4.9±3.3C(n = 25) 1.7 ± 1.9 (n = 25) 2.1 ±0.6C(n = 20) 2.4±0.9C(n = 25) 1.2 ± 0.4 (n = 25) B TFPI Diabetic ESRF Nondiabetic ESRF Age-matched control C Thrombin generation Diabetic ESRF Nondiabctic ESRF Age-matched control reported that the increase in FVIIa during hemodialysis is probably caused by endothe lial damage, resulting in tissue factor expres sion, and that this increase is related to the high prevalence of cardiovascular complica tions. This Finding that the increase in FXIIa in patients with ESRF is not directly related to hyperactivation in the extrinsic pathway is inconsistent with a previously in vitro study , In the case of FXII deficiency during car diac bypass surgery, thrombin generation, di rectly initiated by the formation of tissue factor-FVIIa complexes is observed , Since plasma TFPI was significantly increased in these patients, there appears to be a compen satory mechanism to control hyperactivation caused by FVII activation. 166 Haemostasis 1997;27:163-167 The progression of endothelial damage among patients with ESRF is believed to be more rapid in diabetic than in nondiabetic patients . However, there is no evidence that the hypercoagulability in ESRF patients with well-managed diabetes is more marked than that in nondiabetic patients with ESRF. There were also no differences between these two groups regarding the levels of TFPI. Dia betic patients with ESRF undergoing hemodi alysis did not show characteristic findings of the underlying disease. The TFPI activity in ESRF is reported to be increased due to long term use of heparin . The increase in plas ma TFPI levels is often caused by endothelial dysfunction, which is essentially the same mechanism as that described for plasma thrombomodulin levels. Plasma tissue factor Matsuo/Koide/Kario/Suzuki/Matsuo Downloaded by: Vanderbilt University Library 220.127.116.11 - 10/27/2017 2:04:41 PM a p < 0 .0 5 ,b p < 0.01,'c p < 0.001: compared to controls;d p < 0.05 compared to nondiabetic ESRF. induced by tissue factor expression on the endothelium seems likely to be involved in a marker of endothelial damage and an activa tion signal of the extrinsic pathway . The tissue factor levels in ESRF with and without diabetes showed the same degree of elevation, suggesting that the severity of endothelial damage in ESRF is not affected by the status of the underlying diabetes. Hypercoagulability in ESRF leads to en hanced thrombin generation . The results of this study indicate that the same degree of hypercoagulability was present in both groups with ESRF and that there was no direct con tribution from the increase in FXIIa as there was no correlation between FXIIa and throm bin generation markers. Consistent with these results, no correlation was found between fragment 1+2 production and the level of FXIIa in cardiopulmonary bypass surgery . Thus, this study provided no evidence that the increase in FXIIa is directly related to hyperactivation in the extrinsic pathway in ESRF. Acknowledgments This research was in part funded by a grant from the Heparin Conference (Tokyo, Japan). References Extrinsic Coagulation Factors, FXIIa and ESRF 6 Sandset PM, Larsen ML, Abildgaard U, Lindahl KA, Odcgaard OR: Chromogcnic substrate assay of extrinsic pathway inhibitor (EP1) levels o f the normal population and relation to cholesterol. Blood Coagul Fibrinolysis 1991;2:425-433. 7 Kario K, Matsuo T, Matsuo M, Koide M, Yamada T. Nakamura S, Sakata T, Kato H, Miyata T: Marked increase of activated factor VII in uremic patients. Thromb Haemost 1995;73:763-767. 8 Mitropoulos KA, Martin JC, Stir ling Y, Morrisey JH. Cooper JA: Ac tivation of factors XII and VII in duced in citrated plasma in the pres ence of contact surface. Thromb Haemost 1995;73:763-767. 9 Burman JF, Chung HI, Lane DA, Philippou H, Adami A, Lincoln JCR: Role of factor XII in thrombin generation and fibrinolysis during cardiopulmonary bypass. Lancet 1994;344:1192-1993. 10 Kario K, Matsuo T, Kobayashi H, Matsuo M, Sakata T. MiyataT: Ac tivation of tissue factor-induced co agulation and endothelial cell dys function in non-insulin-dependent diabetic patients with microalbu minuria. Arterioscler Thromb Vase Biol 1995;15:1114-1120. 11 Kario K, Matsuo T, Yamada T, Matsuo M: Increased tissue factor pathway inhibitor levels in uremic patients on regular hemodialysis. Thromb Haemost 1994;71:275— 279. 12 Koyama T, Nishida K: Determina tion of plasma tissue factor antigen and its clinical significance. Br J Haematol 1994;87:343-347. 13 Takagi M, Wada H, Mukai K, Kihara H. Sano S, Minamikawa K, Wakita Y, Nakase T, Nagaya S, Deguchi K, Shirakawa S. Suzuki K: Increased vascular endothelial cell markers in patients with chronic renal failure on maintenance haemodialysis. Blood Coagul Fibrinolysis 1994; 5:713-717. 14 Boisclair MD, Lane DA, Philippou H, Esnouf PE, Sheikh H, Hunt B. Smith KJ: Mechanisms of thrombin generation during surgery and car diopulmonary bvpass. Blood 1993; 82:3350-3357. ' Haemostasis 1997;27:163-167 167 Downloaded by: Vanderbilt University Library 18.104.22.168 - 10/27/2017 2:04:41 PM 1 Kario K, Matsuo T, Yamada T, Nakao K, Shimano C, Matsuo M: Fac tor VII hyperactivity in chronic dial ysis patients. Thromb Res 1992;67: 105-113. 2 Furie B, Furie BC: Molecular and cellular biology of blood coagula tion. N Engl J Med 1992;326:800806. 3 Mitropoulos KA. Martin JC, Bur gess MP, Esnoup MP, Stirling Y, Howarth DJ, Reeves BEA: The in creased rate of activation of factor XII in late pregnancy can contribute to the increased reactivity of factor VII. Thromb Hacmost 1990;63: 349-355. 4 Nakamura S, Kamikubo Y, Okajima K, Asakura H, Nakamura K: Plasma tissue factor, its sensitive as say and characterization (abstract). Thromb Haemost 1993;69:744. 5 Kario K, Miyata T, Sakata T, Mat suo T, Kato H: Fluorogenic assay of activated factor VII: Plasma factor Vila levels in relation to arterial car diovascular disease in Japanese. Arteriosclcr Thromb Vase Biol 1994; 14:265-274.