Tissue Factor and Coagulation Factor VII Levels During Acute Myocardial Infarction
Association With Genotype and Adverse Events
Objective— We investigated in patients with ongoing myocardial infarction (MI) whether coagulation factor VII (FVII) and tissue factor (TF) levels are affected at admission by genetic components and whether they may predict subsequent cardiovascular events.
Methods and Results— 256 patients admitted for MI were evaluated for FVII and TF antigen levels before any treatment at entry, and were genotyped for FVII and TF polymorphisms. FVII gene insertions at −323, 11293 and the −402G/A change predicted FVII levels and explained 14% of variance. The −603 TF gene polymorphism failed to affect significantly TF levels (P=0.07). These variables were correlated with the incidence of death (36 patients) and reinfarction (9 patients) after a median follow-up of 397 days. Events were independently predicted by FVII (HR 2.1, 95% CI 1.2 to 5.7) and TF (HR 4.1, 95% CI 2 to 11) levels. Composite end point was significantly worse when both parameters were above the receiver-operating characteristics (ROC) values (HR 8.3, 95% CI 5 to 18, compared with FVII and TF below), and above the ROC value of TF (>630 pg/mL) it differed among FVII genotype groups.
Conclusions— Admission FVII and TF antigen levels, partially predicted by polymorphisms, are independent predictors of mortality and reinfarction in patients with acute MI.
Coagulation factor VII (FVII) and tissue factor (TF) play a key role in activating the extrinsic coagulation pathway and might be involved, as a single player or in combination, in triggering acute coronary syndromes (ACS). In healthy subjects and in patients with chronic ischemic heart disease, it has been demonstrated that FVII and TF levels are influenced by both environmental and genetic factors.1–4 Previous studies suggested that elevation of FVII may predict coronary events in healthy middle-aged subjects.5–6 However, others studies did not confirm these findings.7–8 In a consecutive group of patients undergoing diagnostic catheterization, TF levels were reported to be higher in those presenting with clinical instability and to predict the need for subsequent revascularization.9 Moreover, in a sub-study of the WARIS-II trial, involving patients with myocardial infarction (MI), increased levels of TF, measured 5 to 7 days after the acute event, were independently associated to the occurrence of future clinical events.10
The role of admission FVII and TF levels in patients with acute MI, however, has been less extensively studied. In particular, very limited data are available regarding the prognostic influence of FVII and TF levels at entry in patients admitted for acute MI. Moreover, whether FVII and TF levels, also during the acute phase of myocardial injury, are influenced by gene polymorphisms is unknown.
The purpose of the present article was to analyze whether FVII and TF levels at admission are affected by intragenic polymorphisms and assess their association to long-term outcome in patients with ongoing MI.
Between May 2003 and June 2004, 256 subjects with MI were enrolled in the present study. Patients were selected from a cohort of 378 patients consecutively referred to our Coronary Care Unit for ACS, based on the following inclusion criteria: prolonged chest pain occurring at rest accompanied by electrocardiographic ischemic changes defined as at least one of the following conditions: ST-segment elevation or depression ≥1 mm, T waves inversion, pseudo-normalization of previously negative T waves. Patients with symptoms onset >24 hours, or immunologic disorder or liver cirrhosis, use of oral anticoagulants or contraceptive were deemed ineligible for the study. The study was approved by the local Ethics Committee. All patients gave written informed consent. The median symptom onset-entry time was 5.3 hours (range 1 to 19 hours). Myocardial necrosis was defined as the elevation of CK-MB and/or Troponin I above the upper reference limit (respectively, 5 ng/mL and 0,1 ng/mL) in 2 or more consecutives samples.11
Blood Sample Collection
Blood withdrawal was performed from all patients on admission before any invasive procedure and the start of any therapy. Then, the samples were centrifuged for 20 minutes at 2000g. All samples were stored at −80°C for DNA extraction and plasma determination.
Factor VII and Tissue Factor Assays
FVII antigen (ng/mL) was assayed in plasma by immunoenzymatic test (Asserachrom VII:ag, Diagnostica Stago). The intra-assay coefficient of variation was 6.9%, and the inter-assays coefficient of variation was 6.9%. Plasma TF antigen (pg/mL) was determined by ELISA method using IMUBIND Tissue Factor ELISA Kit (American Diagnostica Inc). The intra-assay coefficient of variation was 13.2%, and the inter-assays coefficient of variation was 14.7%.
Mutation Analysis and Nomenclature
DNA was extracted from peripheral blood lymphocytes by salting-out method. Three FVII gene polymorphisms were investigated: −402G/A12; −323 decamer insertion (A1 and A2 alleles; A2 defined as presence of the insert)13; the 11293 to 11295insAA in the 3′ untranslated region (D and I alleles, I defined as the presence of the insert).14 Also the TF −603A/G promoter polymorphism, belonging to a specific haplotype with linked polymorphisms at positions −1812, −1332, −1208, was tested.15
Clinical Follow-Up and End Point definitions
Patients underwent outpatient visits every six months and no patient was lost to follow-up.
Our composite end point was the cumulative incidence of death and reinfarction. Reinfarction was diagnosed in the presence of new ischemic symptoms and recurrent elevation of biochemical myocardial necrosis markers (CK-MB and/or Troponin I that were upper the normal range in two or more separate blood samples) with or without concurrent electrocardiographic changes.
Continuous data (normally distributed at Kolmogorov–Smirnov goodness-of-fit test) are presented as means±SD, with the significance of differences judged by t test. Categorical variables were summarized in terms of number and percentages and were compared using two-sided Fisher exact test. Spearman correlation coefficients were used to detect any association between variables. According to previous studies,9–10 it was hypothesized that patients with high (above median value) and low (below median value) levels of TF antigen would display a composite end point free survival rate of 75% and 90% at 1 year, respectively. Therefore, a final population of ≥240 was needed (1-sided α error=0.05, power=0.85). FVII and TF levels were compared among patients with different polymorphisms by analysis of variance, with use of the Tukey procedure for post hoc multivariate comparison of the means. We applied a linear model to evaluate the possible relationship between variables and plasma proteins levels. The prognostic value of variables in Tables 1 and 2⇓ was examined using a Cox-proportional hazards model. Multivariate analysis was performed to identify the independent predictors for adverse events. Survival curves were generated by the Kaplan–Meier method, and survival among groups was compared using the Log Rank test. To compare the predictive role of different values of the studied parameters, receiver operating characteristics (ROC) whit their area under the curve were constructed. The best prognostic cutoff for freedom from composite end point was defined as the one that maximized both sensitivity and specificity. Probability was significant at a level of <0.05. Analysis was performed using STATISTICA 6.1 (Statsoft Inc) and R-language (R Foundation).
Factor VII Plasma Levels and Gene Polymorphisms
FVII plasma levels across the population are presented in Table 2. FVII did not differ in ST-segment elevation myocardial infarction (STEMI) patients as compared with non–ST-segment elevation myocardial infarction (NSTEMI) group (583±193 versus 581±215 ng/mL, P=0.9) and it was not related to any of the variables listed in Table 1. No relationship was found between FVII and CK-MB release or left ventricular ejection fraction (LVEF). Gene frequencies of FVII polymorphisms were in Hardy–Weinberg equilibrium and were similar to those previously reported in the Italian population2 (Table 2). Homozygosis for the A2 allele was associated with 40% lower FVII levels, whereas homozygosis of the A allele of the −402 polymorphism was associated with 40% higher FVII levels (Table 2). 11293 polymorphism was in linkage disequilibrium with −323 polymorphisms and showed similar FVII antigen levels (Table 2). These polymorphisms explained 14% of the total variance of FVII levels (adjusted R2: 0.1368). Figure 1 reports the distribution of FVII levels in patients grouped by −323 and −402 genotypes.
Tissue Factor Plasma Levels and Gene Polymorphism
As for FVII antigen, there was no difference in terms of TF levels between STEMI and NSTEMI groups (577±180 versus 576±118 pg/mL, P=0.9). Age showed a weak correlation with TF levels (r=0.16, P=0.01). Patients with previous MI had higher TF level (620±201 versus 570±150 pg/mL, P=0.05). No relationship was found between TF and CK-MB release or LVEF. Gene frequencies of −603 polymorphism were in Hardy–Weinberg equilibrium and were similar to those previously reported in the European population4 (Table 2). The −603 polymorphism genotypes failed to affect significantly TF levels (P=0.07, Table 2).
After a median follow-up of 397 days (range, 312 to 435 days), 36 deaths (14%) and 9 reinfarction (3.5%) were observed. A total of 45 patients (17.5%) reached the composite end point (35 in the STEMI group and 10 in the NSTEMI group, P=0.2). Those satisfying the composite end point were on average older (73±8 versus 66±12 years, P<0.001), more often had previous MI (33% versus 16%, P=0.04), displayed higher Killip class at entry (31% versus 9%, P=0.002) and lower LVEF (44±10% versus 48±11%, P=0.03). Moreover, they tended to show higher levels of CK-MB, but the difference did not reach statistical significance (170.4±198 versus 140.7±178 ng/mL, P=0.3).
In those patients who reached the composite end point, FVII and TF levels were higher than those with follow up free from adverse events (687±232 versus 560±185 ng/mL, P<0.0001, and 711±206 versus 550±135 pg/mL, P<0.0001). Similar findings were found analyzing the incidence of death alone (673±230 versus 568±191 ng/mL, P=0.003, and 675±115 versus 562±164 pg/mL, P<0.0001).
FVII and TF gene polymorphisms were not significantly related with the clinical outcome. Also considering the genotype groups predicting significant differences in FVII levels (Figure 1), the rate of adverse events did not differ significantly (P=0.4, log-rank test).
At Cox proportional hazards regression, age, LVEF, FVII, and TF antigen levels (evaluated both as single changes unit and as below versus above median value), together with Killip class and previous MI were predictors of composite end point.
By incorporating as putative predictors all significant variables at univariate analysis plus ST-segment elevation at entry, hyperlipidemia and fibrinogen, the independent predictors of composite end point were age, Killip class, FVII antigen, and TF antigen (Figure 2). FVII and TF antigen were also independent predictors of death alone (respectively, HR 1.9, 95% CI 1.3 to 6.7; and HR 3.7, 95% CI 1.6 to 8).
FVII and TF antigen remained associated with composite end point and with death when the analysis was restricted to both STEMI and NSTEMI subgroups.
The cutoff value of FVII at entry for the prediction of composite end point was 618 ng/mL, as identified by ROC. This cutoff had 64% sensitivity and 67% specificity. The area under the ROC curve was 0.68 (95% CI 0.6 to 0.76). The cutoff value of TF was 630 pg/mL. This cutoff had 66% sensitivity and 79% specificity. The area under the ROC curve was 0.77 (95% CI 0.69 to 0.84).
Figure 3 shows differences in composite end point when patients were stratified according to FVII (3A) and TF (3B) levels (under versus above ROC value).
To explore the additive prognostic value of combining FVII and TF levels, Kaplan–Meier curves were constructed according to the combinations generated by having lower or higher identified ROC values (Figure 4A). Composite end point was significantly worse when both parameters were above the ROC values (HR 8.3, 95% CI 5 to 18, compared with patients with low FVII and TF).
Kaplan–Meier curves were constructed also combining the −323 and −402 FVII genotypes with TF levels (above versus below ROC value). Among patients with low TF, FVII genotype influence was negligible (P=0.6, data not shown). Differently, in the high TF group (Figure 4B), there was a significant difference in the composite end point free survival when the −323A2 carriers/−402GG were compared with all homozygotes for the −323A1 (P=0.05) or those who were carriers of the −402A (P=0.05).
Although it is well known that FVII and TF are influenced by environmental and genetic factors,1–4 their circulating levels in patients with ongoing MI and their relationship with environmental and genetic factors are poorly known. Moreover, although previous studies investigated the association between different hemostatic factors and the occurrence of coronary events,5–10 it remains unclear whether TF and FVII levels, measured in patients with MI soon after symptoms onset, may carry valuable prognostic information.
The major findings of our study are:
Plasma protein levels, particularly FVII, are partially influenced by gene polymorphisms even in the first hours of a myocardial necrotic injury, with the 3 investigated FVII gene polymorphisms explaining 14% of the total variance of FVII.
FVII and TF antigen levels, measured at entry, emerged as independent predictors of death and reinfarction.
Outcome was significantly worse when both studied factors were simultaneously elevated, based on the cut-off provided by ROC analysis. FVII genotype influenced outcome only in patients with high TF values.
This is the first study to address the issue whether FVII and TF gene polymorphisms are associated with variations of FVII and TF antigen levels during the first hours of MI (within 24 hours from symptoms onset). With that respect, FVII gene polymorphisms showed a significant although weak influence on FVII plasma levels, whereas TF gene polymorphism failed to affect TF plasma levels.
Previous studies, including patients with stable coronary artery disease, showed that FVII and TF polymorphisms could influence the levels and the activity of these proteins, and as such being associated with a reduced risk of MI.2–4 Our data extend previous findings to the acute setting of MI. Interestingly, when compared with studies performed in healthy volunteers,16 FVII polymorphisms explained a lower proportion of FVII plasma levels. Hypertension, diabetes mellitus, elevated low density lipoproteins, and cigarette smoking, which are known from previous studies to be associated with increased coagulation factor levels, did not show an appreciable influence both on FVII and TF levels during acute MI. In these conditions upregulation of promoters could mask other environmental components, thus possibly explaining our findings.
Our analysis suggests that TF and FVII antigen are independent predictors for the composite end point of death and reinfarction. Importantly, the predictive value of FVII and TF antigens was not restricted to the composite end point but also predicted mortality alone. Moreover, finding that composite end point was significantly worse when both TF and FVII levels were simultaneously above the cut-off provided by ROC analysis further suggests their role to be additive, which is a biologically plausible observation.
The reasons behind the observed association between FVII-TF levels and outcome may lie in their influence in the size and stability of the thrombus, and hence in the consequent myocardial damage. Interestingly, in our study no relationship was found between FVII-TF levels and myocardial damage evaluated both as peak of CK-MB release or LVEF. Moreover, the FVII-activated TF complex may also play a role in the migration and proliferation of vascular smooth muscle cells,17 in vascular remodeling, and in plaque neovascularization18 and thereby in promoting plaque destabilization.
Prior studies have reported conflicting results. In a study by Malarstig and colleagues involving ACS patients, no association between plasma TF and outcome was observed.19 Other studies reported TF levels to be an independent predictor of cardiovascular events.9–10 The patient selection (ongoing MI versus elective patients) and the different sampling schedule (first hours versus days or weeks after symptoms onset) could contribute to explain our results in the context of previous findings.
As previously reported,20 no relationship was found in our population between −603 A/G TF gene polymorphism and cardiovascular events. FVII genotypes predicted FVII levels but they were significantly associated to the outcome only in the high risk subgroup of patients characterized by having elevated TF (above ROC value). However, it should be noted that the limited sample size of our study does not allow drawing definitive conclusions in this regard. Bearing in mind these limitations, our data suggest that at admission circulating levels of FVII and TF are more predictive than their polymorphisms, thus supporting the role of other genetic and/or acquired factors in the regulation of the expression of these genes.
We have not measured the activity of FVII and TF forms. Both FVII and TF are present in plasma in different molecular forms reflecting activation or exposure on membranes. The analysis of FVII and TF antigen reflects total FVII, both zymogene and activated forms, and total TF, both membrane-bound and soluble forms. Although we are well aware that the various FVII and TF forms could play different roles in thrombus formation, we assumed, in accordance with other authors,10,21 that elevation of FVII and TF antigen reflects their procoagulant activity. Recently, it has been suggested that the TF determination with a commercially available immunoassay is not satisfactory.19,22 Nevertheless, this immunoassay has been extensively used in previous studies10,19 and our data are in the range of values obtained with the new improved assay.22 The sample size of our study was too small to allow an affordable estimate of predictive role of gene polymorphisms. In particular, to obtain a reliable estimate of the prognostic capability of gene polymorphisms and if they could be a simple method to stratify the patients, a larger prospectively collected study population is clearly in demand. Our study was powered to evaluate the role of TF in predicting outcome and all other analyses performed, particularly the prognostic influence of gene polymorphisms and the causal relationship between environmental factors-plasma proteins-myocardial damage-cardiac function, should be considered exploratory and hypothesis-generating. Additional clinical investigations are also needed to understand whether patients have elevated plasma levels also during hospitalization and follow-up. This could allow to understand the influence of acute event on levels variations and to select the ideal timing to identify patients with a blood prone to thrombosis and thus at high risk of adverse events. Our data may suggest that samples obtained at entry are suitable for such a purpose.
Admission FVII and TF antigen, partially predicted by polymorphisms, were independent predictors of composite end point (mortality and reinfarction), which was even worse in patients showing both parameters simultaneously elevated. FVII genotype influenced outcome only in patients with high TF values. Our findings may reinforce the interest to further evaluate, in the clinical setting, the benefit of using drugs able to interfere with the action of FVII-TF complex, which are currently being developed.
Sources of Funding
This study was supported by grants from Fondazione Cassa di Risparmio di Ferrara, by ProgettoARTGEA-Comitato dei Sostenitori, and “Programmi Ricerca Cofinanziati” MIUR.
Original received June 19, 2006; final version accepted September 5, 2006.
Mariani G, Bernardi F, Bertina R, Vicente V, Prydz H, Samama M, Sandset PM, Di Nucci GD, Testa MG, Bendz B, Chiarotti F, Ciarla M, Strom R. Serum phospholipids are the main environmental determinants of activated factor VII in the most common FVII genotype. Haematologica. 1999; 84: 620–626.
Heinrich J, Balleisen L, Schulte H, Assman G, van de Loo J. Fibrinogen and Factor VII in the prediction of coronary risk: results from the PROCAM study in healthy men. Arterioscler Thromb. 1994; 14: 1392–1399.
Folsom AR, Wu KK, Rosamond WD, Sharrett AR, Chambless LE. Prospective study of hemostatic factors and incidence of coronary heart disease: the Atherosclerosis Risk in Communities (ARIC) Study. Circulation. 1997; 96: 1102–1108.
Smith FB, Lee AJ, Fowkes FG, Price JF, Rumley A, Lowe GD. Hemostatic factors as predictors of ischemic heart disease and stroke in the Edinburg Artery Study. Arterioscler Thromb Vasc Biol. 1997; 17: 3321–3325.
Soejima H, Ogawa H, Yasue H, Kaikita K, Nishiyama K, Misumi K, Takazoe K, Miyao Y, Yoshimura M, Kugiyama K, Nakamura S, Tsuji I, Kumeda K. Heightened tissue factor associated with tissue factor pathway inhibitor and prognosis in patients with unstable angina. Circulation. 1999; 99: 2908–2913.
Myocardial infarction redefined—a consensus document of the Joint European Society of Cardiology/American College of Cardiology Committee for the Redefinition of Myocardial Infarction. Eur Heart J. 2000; 21: 1502–1513.
van ’t Hooft FM, Silveira A, Tornvall P, Iliadou A, Ehrenborg E, Eriksson P, Hamsten A. Two common functional polymorphisms in the promoter region of the coagulation factor VII gene determining plasma factor VII activity and mass concentration. Blood. 1999; 93: 3432–3441.
Bernardi F, Arcieri P, Bertina RM, Chiarotti F, Corral J, Pinotti M, Prydz H, Samama M, Sandset PM, Strom R, Garcia VV, Mariani G. Contribution of factor VII genotype to activated FVII levels. Differences in genotype frequencies between northern and southern European populations. Arterioscler Thromb Vasc Biol. 1997; 17: 2548–2553.
Sabater-Lleal M, Martinez-Marchan E, Martinez-Sanchez E, Coll M, Vallve C, Mateo J, Souto JC, Fontcuberta J, Soria JM. Complexity of the genetic contribution to factor VII deficiency in two Spanish families: clinical and biological implications. Haematologica. 2003; 88: 906–913.
Cirillo P, Cali G, Golino P, Calabro P, Forte L, De Rosa S, Pacileo M, Ragni M, Scopacasa F, Nitsch L, Chiariello M. Tissue factor binding of activated factor VII triggers smooth muscle cell proliferation via extracellular signal-regulated kinase activation. Circulation. 2004; 109: 2911–2916.
Rao VM, Pendurthi U. Tissue factor-factor VIIa signaling. Arterioscler Thromb Vasc Biol. 2005; 25: 47–56.
Malarstig A, Tenno T, Johnston N, Lagerqvist B, Axelsson T, Syvanen A, Wallentin L, Siegbahn A. Genetic variations in the tissue factor gene are associated with clinical outcome in acute coronary syndrome and expression levels in human monocytes. Arterioscler Thromb Vasc Biol. 2005; 25: 2667–2672.
Arnaud E, Barbalat V, Nicaud V, Cambien F, Evans A, Morrison C, Arveiler D, Luc G, Ruidavets JB, Emmerich J, Fiessinger JN, Aiach M. Polymorphisms in the 5′ regulatory region of the tissue factor gene and the risk of myocardial infarction and venous thromboembolism: the ECTIM and PATHROS studies. Arterioscler Thromb Vasc Biol. 2000; 20: 892–898.
Soejima H, Ogawa H, Yasue H, Kaikita K, Takazoe K, Nishiyama K, Misumi K, Miyamoto S, Yoshimura M, Kugiyama K, Nakamura S, Tsuji I. Angiotensin-converting enzyme inhibition reduces monocyte chemoattractant protein-1 and tissue factor levels in patients with myocardial infarction. J Am Coll Cardiol. 1999; 34: 983–988.