activatable fibrinolysis inhibitor (TAFI) levels
N. Folkeringa M. Coppens
6
6 Absolute risk of venous and arterial thrombo embolism in thrombophilic families is not increased by high
thrombin-activatable fibrinolysis inhibitor
Abstract
High levels of thrombin-activatable fibrinolysis inhibitor (TAFI) are a supposed risk factor for thrombosis. However, results from previous studies are conflicting.We as-sessed the absolute risk of venous and arterial thromboembolism in subjects with high TAFI levels (>126 U/dl) versus subjects with normal levels, and the contribution of other concomitant thrombophilic defects. Relatives from four identical cohort studies in families with either deficiencies of antithrombin, protein C or protein S, prothrombin 20210A, high factor VIII levels, or hyperhomocysteinemia were pooled.
Probands were excluded. Of 1,940 relatives, 187 had high TAFI levels. Annual inci-dences of venous thromboembolism were 0.23% in relatives with high TAFI levels versus 0.26% in relatives with normal TAFI levels (adjusted relative risk [RR] 0.8;
95% confidence interval [CI], 0.5–1.3). For arterial thrombosis these were 0.31% ver-sus 0.23% (adjusted RR 1.4; 95% CI, 0.9–2.2). High levels of factor VIII, IX and XI were observed more frequently in relatives with high TAFI levels. Only high factor VIII lev-els were associated with an increased risk of venous and arterial thrombosis, inde-pendently of TAFI levels. None of these concomitant defects showed interaction with high TAFI levels. High TAFI levels were not associated with an increased risk of ve-nous and arterial thromboembolism in thrombophilic families.
Introduction
Thrombin-activatable fibrinolysis inhibitor (TAFI) is a procarboxypeptidase which suppresses fibrinolysis by removing carboxy-terminal lysine residues from partially degraded fibrin.1-3These residues are involved in binding of plasminogen and tis-sue-type plasminogen activator, and in plasmin formation. TAFI is activated by thrombin, mainly in complex with thrombomodulin, and by plasmin.4Since acti-vated TAFI (TAFIa) inhibits fibrinolysis,5high plasma levels of TAFI may promote the development of thrombosis. Of six previous studies, four reported an association of first or recurrent venous thromboembolism with high TAFI levels,6-9while two studies did not.10,11The results of studies on the risk of arterial thromboembolism related to high TAFI levels, were also not consistent.12-23
Venous thrombosis is considered a multicausal disease. It has been demon-strated that aggregation of thrombophilic defects in families with hereditary defi-ciencies of antithrombin, protein C, or protein S increases the risk of venous thromboembolism.24-27A similar interaction has been suggested for concomitance of high levels of factors VIII, IX and XI in subjects with high TAFI levels.7,9However, absolute risks of venous and arterial thromboembolism for different combinations of high TAFI levels with other thrombophilic defects are unknown.
We performed a study to assess the absolute risk of venous and arterial throm-boembolism in subjects with high TAFI levels, and the contribution of concomitant thrombophilic defects.
Materials and Methods
Study population
We pooled subjects from four previous family cohort studies with an identical de-sign.28-31In a family cohort study on an inherited risk factor, symptomatic probands with this risk factor are used to identify subjects (relatives with the risk factor, whether or not symptomatic) and controls (relatives without the risk factor, whether or not symptomatic). Absolute risks in subjects and controls can be estimated be-cause the duration of exposure to the risk factor is known. The four studies were per-formed to estimate the absolute risk of venous thromboembolism, associated with hereditary deficiencies of antithrombin, protein C or protein S,27,31the prothrombin 20210A mutation,28 elevated plasma levels of factor VIII:C29 and hyper homo -cysteinemia,30respectively. Briefly, probands in each of these studies were conse -cutive patients with documented venous thromboembolism, or premature athero sclerosis (age <50 years), and one of the above mentioned thrombophilic de-fects. First-degree relatives, and for the rare antithrombin deficiency also second-degree relatives from a deficient parent, who were 15 years of age or older were High TAFI levels and risk of thromboembolism
identified by pedigree analysis. They were enrolled after written informed consent was obtained. Detailed information about previous episodes of venous and arterial thromboembolism, exposure to external risk factors for venous thrombosis, risk fac-tors for atherosclerosis, and anticoagulant prophylaxis or treatment was collected by the physicians of the thrombosis out-patient clinics of the three participating hos-pitals using a validated questionnaire32and reviewing medical records. In relatives, blood samples for thrombophilia testing were taken after clinical data had been col-lected. They were tested for all abovementioned thrombophilic defects, as well as fac-tor V Leiden and increased levels of facfac-tor IX:C and facfac-tor XI:C. In the present study, TAFI levels were measured additionally in relatives who consented. All studies were approved by the institutional review boards of the participating hospitals.
Definitions
Venous thromboembolism was considered established if deep vein thrombosis (DVT) was confirmed by compression ultrasound or venography, and pulmonary embolism by ventilation/perfusion lung scanning, spiral computed tomography (CT) scanning or pulmonary angiography, or when the patient had received full-dose unfraction-ated heparin and a vitamin K antagonist for at least three months without objective testing at a time when these techniques were not available yet. Venous thromboem-bolism was classified as provoked if it had occurred at or within three months after exposure to one or more external risk factors, including surgery, trauma, immobi-lization for more than seven days, use of oral contraceptives or hormonal replace-ment therapy, pregnancy or puerperium, and malignancy. In the absence of these risk factors, it was considered spontaneous.
Coronary and peripheral artery disease had to be symptomatic and angiographi-cally proven. Myocardial infarction had to be diagnosed in accordance with clinical, enzymatic and electrocardiographic criteria. Ischemic stroke was defined as the onset of rapidly developed symptoms and signs of loss of cerebral function, which lasted at least 24 hours (h) and had an apparent vascular cause, as demonstrated by CT or mag-netic resonance imaging (MRI). If a cerebral event completely resolved within 24h, it was classified as transient ischemic attack (TIA). Risk factors for atherosclerosis in-cluded diabetes mellitus, smoking, hyperlipidemia and hypertension.
Laboratorystudies
Plasma levels of activatable TAFI were measured by a two-step chromogenic syn-thetic substrate assay (Pefakit® TAFI, Pentafarm, Basel, Switzerland). In this assay TAFI is first activated by the thrombin/thrombomodulin complex followed by cleav-age of a thio-derivated arginine substrate, which is monitored by color development after reaction with Ellman’s reagent (5,5’-dithiobis-2-nitrobenzoic acid) . The assay was calibrated with pooled normal plasma from 47 healthy volunteers (age 22–57 years; 40% women, not using oral contraceptives) set at 100 U/dl. The normal range was determined with 40 individual plasmas from healthy volunteers, who had no
(family) history of venous thromboembolism and were neither pregnant, nor had used oral contraceptives for at least three months. The estimated range was 79–
126 U/dl, after exclusion of two outlyers, 160 U/dl and 163 U/dl, respectively. Dilu-tion in TAFI-deficient plasma (Affinity Biologicals, Ancaster, Canada) in stead of water did not increase the recovery of TAFI (98% vs. 106%), which confirmed the claimed specificity of the assay for TAFI. In addition, this assay previously showed no interference with other carboxypeptidases and uses a short reaction time (5 minutes [min]) after activation of TAFI to overcome the relative instability of TAFIa.33TAFI levels were defined as high above 126 U/dl. Activity of antithrombin (Coatest®, Chro-High TAFI levels and risk of thromboembolism
Table 1. Characteristics of relatives with high and normal TAFI levels.
High TAFI levels Normal TAFI levels
(n=187) (n=1,753) P
TAFI level, IU/dl, median (range) 133 (127–184) 98 (19–126)
Women, n (%) 81 (43) 790 (45) 0.70
Venous thromboembolism, n (%) 15 (8) 131 (7) 0.77
Provoked, n (%) 5 (33) 49 (37)
Spontaneous, n (%) 10 (67) 82 (63)
Age at onset, years, mean (SD) 39 (16) 44 (18) 0.29
Arterial thromboembolism, n (%) 20 (11) 120 (7) 0.07
Ischemic stroke /TIA, n (%) 3 (1.6) 50 (2.9)
Myocardial infarct, n (%) 9 (4.8) 43 (2.5)
Peripheral arterialobstructive disease, n (%) 8 (4.3) 27 (1.5)
Age at onset, years, mean (SD) 62 (13) 61 (14) 0.84
Risk factors
Hypertension, n (%) 60 (32) 314 (18) <0.001
Hyperlipidemia, n (%) 51 (27) 207 (12) <0.001
Smoking, n (%) 67 (36) 615 (36) 0.94
Diabetes mellitus, n (%) 13 (7) 76 (4) 0.14
Thrombophilic defects, % (n tested)
Antithrombin deficiency 4 (186) 3 (1740) 0.35
Protein C deficiency 3 (186) 4 (1739) 0.84
Protein S deficiency type I 5 (183) 3 (1722) 0.29
Protein S deficiency type III 9 (176) 10 (1695) 0.51
Prothrombin G20210A 13 (187) 13 (1743) 0.80
Factor V Leiden 14 (187) 13 (1741) 0.77
Factor VIII >150% 40 (186) 30 (1712) 0.003
Factor IX >150% 30 (175) 14 (1659) <0.001
Factor XI >150% 17 (185) 6 (1728) <0.001
Hyperhomocysteinemia 28 (169) 28 (1517) 0.93
TIA, transient ischemic attack.
mogenix, Mölndal, Sweden) and protein C (Berichrom Protein C®, Dade Behring, Marburg, Germany) were measured by chromogenic substrate assays and protein C antigen and total and free protein S antigen levels by Enzyme Linked Immuno Sor-bent Assay (ELISA) (DAKO, Glostrup, Denmark). Antithrombin deficiency was de-fined by lowered levels, i.e. levels below the lower limit of its normal range (<74 U/dl), protein C deficiency type I and type II were defined by lowered levels of protein C anti-gen and/or protein C activity, respectively (63 U/dl). Protein S deficiency type I was defined by lowered total (<67U/dl) and free protein S antigen levels (<65 U/dl), and protein S deficiency type III by normal total protein S antigen and lowered free pro-tein S antigen levels. Deficiencies were considered inherited if they were confirmed by measuring a second sample that was collected three months later and were demonstrated in at least two family members. If there was a discrepancy between the results of the two tests, a third sample was tested. Relatives with acquired ditions were excluded. A deficiency was considered acquired, through use of oral con-traceptives or pregnancy, unless it was confirmed at least three months after withdrawal of oral contraceptives, or delivery, respectively. Relatives with abnormal liver function tests were only evaluable if deficiencies were established at repeated measurements after recovery.
Factor V Leiden and the prothrombin G20210A mutation were demonstrated by polymerase chain reactions.34, 35Factors VIII:C, IX:C and XI:C were measured by one-stage clotting assays and were defined as high at levels above 150 IU/dl.29, 36,37 Fasting levels of homocysteine were measured by high performance liquid chroma -tography.38Hyperhomocysteinemia was defined as a fasting level above 18.5 µM, as described in the Dutch population.39
In probands and relatives with venous thromboembolism, blood samples were collected at least three months after this event had occurred. If they were still treated with acenocoumarol, a short acting vitamin K antagonist, samples were taken after this therapy had been interrupted for at least two weeks, meanwhile nadroparin was given subcutaneously.
Statisticalanalysis
We compared the absolute risk of the first episode of either venous thrombo -embolism or arterial thrombo-embolism in relatives with high TAFI levels and rela-tives with normal TAFI levels. Probands were excluded from analysis to avoid bias.
Annual incidences were calculated by dividing the number of symptomatic relatives by the total number of observation years. When calculating the annual incidence of venous thromboembolism, the occurrence of arterial thromboembolism was ignored and vice versa. Observation time was defined as the period from age 15 years until the first venous or arterial thromboembolic episode, or until the end of study. The ef-fects of concomitant thrombophilic deef-fects on the absolute risks of venous and arte-rial thromboembolism were assessed in relatives with high TAFI levels and relatives with normal TAFI levels, comparing relatives with and those without other
throm-bophilic defects. The relative risks of venous and arterial thromboembolism were calculated from annual incidences. The 95% confidence intervals (95% CI) and p-values were computed using small sample statistics based on mid-P-value functions from the binomial probability model.40Multivariable analysis was performed by Cox proportional hazard regression analysis to adjust the relative risk of thrombosis for variables that showed an association at a p-value <0.20 at univariate analysis. Con-tinuous variables were expressed as mean values and standard deviations and cate-gorical data as counts and percentages. Differences between groups were evaluated by the Student t-test or Mann-Whitney U test, depending on the normality of data for continuous data, and by Fisher exact test for categorical data. A two-tailed p-value of less than 0.05 was considered to indicate statistical significance. Statistical analy-ses were performed using SAS software, version 9.1 (SAS-Institute inc., Cary, NC, USA).
Results
Overall, 1,940 relatives were eligible (age 15 years or older and alive). Their probands had a deficiency of either antithrombin, protein C or protein S (600 relatives), carried the prothrombin 20210A mutation (360 relatives), had high plasma levels of factor VIII:C (543 relatives), or had hyperhomocysteinemia (437 relatives). Of these 1,940 rel-atives, 187 (10%) had high TAFI levels. Their characteristics are listed in Table 1. Men and women were equally distributed among both groups. Fifteen relatives (8%) with high TAFI levels had experienced venous thromboembolism, compared to 131 rela-tives (7%) with normal TAFI levels (p= 0.77). Age at onset of venous thromboembolism was comparable (p=0.29). Arterial thromboembolism had occurred in 20 relatives (11%) with high TAFI levels and in 120 relatives (7%) with normal TAFI levels (p=0.07).
Age at onset of arterial thromboembolism was comparable in both groups (p=0.84).
Of risk factors for atherosclerosis, hypertension (p<0.001) and hyperlipidemia (p<0.001) were more frequently observed in relatives with increased TAFI levels. High levels of factor VIII (p=0.003), factor IX (p<0.001) and factor XI (p<0.001) were more fre-quently observed in relatives with increased TAFI levels. Other thrombophilic defects were equally distributed among both groups.
Annual incidences of venous thromboembolism were 0.23% (95%CI, 0.13–0.38) in relatives with high TAFI levels and 0.26% (95%CI, 0.22–0.31) in relatives with normal TAFI levels. Adjusted for coexistence of other thrombophilic defects, the relative risk (RR) was 0.8 (95%CI, 0.5–1.3) (Table 2). For arterial thromboembolism annual inci-dences were 0.31% (95%CI, 0.19–0.47) in relatives with high TAFI levels and 0.23%
(0.19–0.28) in relatives with normal TAFI levels; RR 1.4 (95% CI, 0.9–2.2), adjusted for gender and classical risk factors for atherosclerosis. Age did not affect the relative risk of venous and arterial thrombo embolism (data not shown).
High TAFI levels and risk of thromboembolism
Cumulative distribution curves (Figure 1) did not show a relationship between ve-nous or arterial thrombotic events and TAFI levels. Comparing quartiles of TAFI lev-els, there were no differences in frequency of venous (p=0.77) and arterial thrombo embolism (p=0.24).
As high plasma levels of factors VIII, IX and XI were more frequently observed in relatives with high TAFI levels than in relatives with normal TAFI levels, the effects of these thrombophilic defects on the risk of thromboembolism were assessed (Table 3). Compared to relatives with normal levels of TAFI and factor VIII (annual incidence 0.17%), annual incidences of venous thromboembolism were 0.11 in relatives with high TAFI levels and normal factor VIII levels (RR 0.6; 95%CI, 0.2–1.5); 0.41% in rel-atives with normal TAFI levels and high factor VIII levels (RR 2.4; 95%CI, 1.7–3.4);
and, 0.41% in relatives with high levels of TAFI and factor VIII (RR 2.4; 95%CI, 1.2–
4.4). High levels of either factor IX or factor XI did not influence the absolute risk in relatives with normal or high TAFI levels. The results were similar for arterial throm-boembolism.
Table 2. Absolute risk of venous and arterial thromboembolism in relatives with high and normal TAFI levels.
High TAFI levels Normal TAFI levels
Venous thromboembolism, n 15 131
Observation period, years 6530 50588
Annual incidence, % (95% CI) 0.23 (0.13–0.38) 0.26 (0.22–0.31)
Crude relative risk (95% CI) 0.9 (0.5–1.5)
Adjusted relative risk (95% CI)* 0.8 (0.5–1.3)
Arterial thromboembolism, n 20 120
Observation period, years 6542 51390
Annual incidence, % (95% CI) 0.31 (0.19–0.47) 0.23 (0.19–0.28)
Crude relative risk (95% CI) 1.3 (0.80–2.07)
Adjusted relative risk (95% CI)# 1.4 (0.9–2.2)
*Adjusted for coexistence of other thrombophilic defects. # Adjusted for gender and classical risk fac-tors for atherosclerosis. CI, confidence interval.
Discussion
Our study did not show a relationship between levels of activatable TAFI and the ab-solute risk of venous thromboembolism. In fact, annual incidences of venous throm-boembolism in both relatives with high (>126 U/dL) TAFI levels (0.23%) and relatives with normal TAFI levels (0.26%) were comparable to its annual incidence in the nor-mal population (i.e. 0.1% - 0.3%).41, 42This finding seems to be in contrast with the re-sults of previous studies.6-11 However, only two of six previous studies provided evidence that high TAFI levels were associated with an increased risk of the first episode of venous thromboembolism.6,7Verdu et al. demonstrated a 4.0– fold higher risk (95% CI, 1.4–10.9) in subjects with TAFI levels >10.9 µg/ml (90thpercentile), com-pared to subjects with lower TAFI levels.6Van Tilburg et al. demonstrated a 1.7-fold higher risk (95% CI, 1.1–2.5) comparing subjects with TAFI levels >122 U/dl (90th per-centile) to subjects who had lower TAFI levels.7In the latter study population, Mar-tini et al. found a 1.3-fold higher risk (95% CI, 1.0–1.6) when they compared carriers and non-carriers of the 505G allele, a polymorphism of the TAFI gene.10Remarkably, TAFI levels were lower in carriers than in non-carriers. Libourel et al. reported a 1.8-fold higher risk (95% CI, 0.6–5.3) of venous thromboembolism in carriers of factor V Leiden who had TAFI levels >115 U/dl, compared to their relatives who only carried factor V Leiden.8Schroeder et al. showed no difference in TAFI levels between pa-tients with pulmonary embolism and controls.11Finally, Eichinger et al. found in a prospective study a 1.7-fold (95% CI, 1.1–2.7) increased risk of recurrent venous throm-boembolism in patients with TAFI levels above the 75thpercentile (men, 114 U/dl and women, 107 U/dl) compared to patients with lower levels.9
High TAFI levels in our study tended to be associated with a 1.4-fold higher risk of arterial thromboembolism. Similar to venous thromboembolism, annual inci-dences of arterial thromboembolism in relatives with high TAFI levels and in rela-tives with normal TAFI levels were, however, within the range reported from the normal population (i.e. 0.1%-0.4%).43The results of previous (case control) studies were not consistent. Acute coronary syndrome was associated with high TAFI levels in two studies,12,13with low TAFI levels in two studies,14,15whereas two studies did not show differences in TAFI levels between cases and controls.16,17Another study reported lower TAFI levels in patients with refractory angina pectoris.18Finally, one study of patients with coronary artery disease, showed an association of high TAFI levels with refractory angina pectoris, compared to non-refractory angina pectoris.19 To the contrary, all four case control studies on stroke did show a higher risk of stroke in patients with high TAFI levels, compared to patients with normal TAFI levels.20–23
Inconsistency of results from abovementioned studies may be due to the wide variation of applied TAFI assays and the difficulty of measuring TAFI levels related to the intrinsic instability of TAFIa associated with a conformational change, which may be followed by proteo lytic inactivation, as well as many polymorphisms in the TAFI gene affecting both protein level and activity. Antigen assays were used in 11 of 16 High TAFI levels and risk of thromboembolism
Figure 1.
Cumulative distribution curves of venous thromboembolism (VTE) (A) and arterial thrombo -embolism (ATE) (B) related to TAFI levels. The vertical dashed line represents the 95th per-centile of TAFI levels. Venous thrombo embolism (p=0.77) and arterial thromboembolism (p=0.23) were equally distributed among quartiles (horizontal dashed lines).
studies6,7,9–11,13–15,17,18,20activity assays in five studies,8,9,12,21,22and both assays in two studies.16,23Furthermore, different assays of either TAFI activity and TAFI antigen were used. High TAFI levels were identified as a risk factor for venous or arterial thrombosis in six of seven studies by activity assays8,9,12,21-23compared to six of 13 studies using antigen assays.6,7,9,13,20,23Two studies, which did not reveal an asso-ciation of specific TAFI polymorphisms with thrombosis, did show an assoasso-ciation of the risk of thrombosis with high TAFI levels.22,23Inaccuracy of TAFI antigen assays may be explained by insensitivity of some ELISA’s for some isoforms of TAFI.44
Measuring levels of activatable TAFI in the present study, we were unable to demonstrate an increased risk of thrombosis in individuals with high TAFI levels.
With the assay of activatable TAFI used in this study, 99% of the tested relatives had levels between 60 and 160 U/dl (Figure 2). Such a range is quite comparable with that of most clotting and anticoagulant factors and indicates a limited effect of both poly-morphisms in the TAFI gene and instability of activated TAFI on our assay results.
The limited number of very high levels of activatable TAFI observed both in the study group (up to 184 U/dl) and healthy volunteers (up to 163 U/dl) might be related to the acute phase nature of TAFI,45,46although this was not supported by the values of other factors measured in the same plasma sample. The few relatives with very low levels of activatable TAFI (as low as 19 U/dl) may be due to chance or heterozygous TAFI deficiency. Interestingly, extreme low and high TAFI levels were observed in rel-atives without thrombosis.
High TAFI levels and risk of thromboembolism
Table 3. Risk of venous and arterial thromboembolism related to levels of TAFI and factor VIII, factor IX and factor XI.
Venous thromboembolism Arterial thromboembolism
Relatives Events Events
(n) (n) Annual incidences Relative risk (n) Annual incidences Relative risk Normal TAFI, normal FVIII 1204 56 0.17 (0.13-0.22) Reference 62 0.19 (0.15-0.24) Reference High TAFI, normal FVIII 111 4 0.11 (0.03-0.27) 0.6 (0.2–1.5) 9 0.24 (0.11-0.64) 1.3 (0.6-2.4) Normal TAFI, high FVIII 508 70 0.41 (0.32-0.52) 2.4 (1.7-3.4) 54 0.30 (0.23-0.40) 1.6 (1.1-2.3) High TAFI, high FVIII 75 11 2.4 (1.2-4.4) 0.41 (0.20-0.73) 11 0.40 (0.20-0.70) 2.1 (1.1-3.9) Normal TAFI, normal FIX 1431 79 Reference 0.20 (0.15-0.24) 89 0.22 (0.18-0.27) Reference High TAFI, normal FIX 122 4 0.09 (0.03-0.24) 0.5 (0.2-1.2) 13 0.31 (0.10-0.53) 1.4 (0.8-2.5) Normal TAFI, high FIX 228 15 0.20 (0.11-0.32) 1.0 (0.6-1.7) 23 0.30 (0.19-0.44) 1.4 (0.8-2.1) High TAFI, high FIX 53 5 0.26 (0.08-0.60) 1.3 (0.5-3.0) 5 0.25 (0.08-0.59) 1.2 (0.4-2.7) Normal TAFI, normal FXI 1625 116 0.25 (0.21-0.30) Reference 118 0.24 (0.20-0.29) Reference High TAFI, normal FXI 153 15 0.7 (0.3-1.3) 0.17 (0.08-0.33) 9 0.29 (0.16-0.48) 1.2 (0.7-2.0) Normal TAFI, high FXI 32 2 1.5 (0.8-2.6) 0.37 (0.18-0.66) 11 0.06 (0.01-0.23) 0.3 (0.04-0.9) High TAFI, high FXI 103 5 1.9 (0.8-4.1) 0.48 (0.18-1.05) 6 0.37 (0.12-0.78) 1.5 (0.6-3.5)
Two previous studies reported a multiplicative interaction of high factor VIII levels and high TAFI levels with respect to the risk of first7and recurrent9venous throm-bo emthrom-bolism, respectively. The latter study also showed an interaction between high levels of factor IX and factor XI in cases with high TAFI levels.9We analyzed possi-ble interactions of high levels of factor VIII, factor IX and factor XI with high TAFI lev-els, considering that concomitance of high levels of factor VIII, factor IX and factor XI were more frequently observed in relatives with high TAFI levels than in relatives with normal TAFI levels, while other thrombophilic defects were equally distributed.
However, subgroup analysis did not show such an interaction, considering that high TAFI levels were not a risk factor for thrombosis. Although relatives with high levels of TAFI and factor VIII were at higher risk of venous and arterial thrombosis, re-spectively, compared to relatives with normal levels of TAFI and factor VIII, this ad-ditional effect was independent of TAFI levels and could be attributed to high factor VIII levels as such.
Some aspect of this retrospective study warrants comment. Probands in our study were consecutive patients with thromboembolism and a thrombophilic defect. Be-cause high TAFI levels were not defined as an index thrombophilic defect, we might have excluded probands in whom thromboembolism was associated with high TAFI levels, and consequently their relatives. However, it is not plausible that TAFI levels are lower in patients with other thrombophilic defects. Moreover, the range of TAFI Figure 2. Boxplot of TAFI levels in the study population (n=1,940).
levels observed in our relatives and the cut-off level of high TAFI levels were compa-rable to their values in other (case control) studies. Because of the retrospective design of our study, TAFI levels were not measured at the time of thrombosis. It is unlikely that this has influenced our results as we found no correlation between age and TAFI levels at time of enrollment (r2=0.038, data not further shown). Referral bias, due to the setting of university hospitals, seems also unlikely to explain our finding that high TAFI levels are not a risk factor for thromboembolism. Venous thromboem-bolism was not established by objective techniques in a number of patients, because these were not available yet. This might have resulted in an overestimated risk of ve-nous thromboembolism. However, the absolute risk would be equally increased in relatives with high TAFI levels and in relatives with normal TAFI levels, while it in fact did not exceed the risk in the normal population.
In conclusion, high TAFI levels did not increase the risk of venous or arterial thromboembolism in this large cohort of thrombophilic families.
References
1. Hendriks D, ScharpeS, van Sande M, et al. Characterisation of a carboxypeptidase in human serum distinct from carboxypeptidase N. J Clin ChemClin Biochem1989;27: 277–285.
2. Eaton DL, Malloy BE, TsaiSP, et al. Isolation, molecular cloning, and partial characterization of a novel carboxypeptidaseB from human plasma. J BiolChem 1991; 266:21:833–21:838.
3. Bajzar L, Manuel R, Nesheim ME. Purification and characterization of TAFI, a thrombin-activatable fibrinolysis inhibitor. J Biol Chem 1995; 270: 14477–14484.
4. Bouma BN, Mosnier LO, Meijers JC, Griffin JH. Factor XI dependent and independent activation of thrombinactivatable fibrinolysis inhibitor (TAFI) in plasma associated with clotformation. Thromb Haemost 1999;82: 1703–1708.
5. Wang W, Boffa MB, Bajzar L, et al. A study of the mechanism of inhibition of fibrinolysis by activated thrombin-activable fibrinolysis inhibitor. J BiolChem 1998; 273:27176–27181.
6. Verdú J, Marco P, Benlloch S, et al. Thrombinactivatable fibrinolysis inhibitor (TAFI) polymorphisms and plasma TAFI levels measured with an ELISA insensitive to isoforms in patients with venous thromboembolic disease (VTD).Thromb Haemost 2006;95: 585–586.
7. vanTilburg NH, Rosendaal FR, Bertina RM. Thrombin activatable fibrinolysis inhibitor and the risk for deepvein thrombosis. Blood 2000;95: 2855–2859.
8. Libourel EJ, BankI, Meinardi JR, et al. Co-segregation of thrombophilic disorders in factor V Leiden carriers; the contributions of factor VIII, factor XI, thrombin activatable fibrinolysis inhibitor and lipoprotein(a) to the absolute risk of venous thromboembolism. Haematologica 2002; 87: 1068–
1073.
9. Eichinger S, Schonauer V, Weltermann A, et al. Thrombin-activatable fibrinolysis inhibitor and the risk for recurrent venous thromboembolism. Blood 2004; 103:3773–3776.
10. Martini CH, Brandts A, de Bruijne EL, et al. The effect of genetic variants in the thrombin activatable fibrinolysis inhibitor (TAFI) gene on TAFI-antigen levels, clot lysistime and the risk of venous thrombosis. Br J Haematol 2006; 134:92–94.
High TAFI levels and risk of thromboembolism
11. Schroeder V, Kucher N, Kohler HP. Role of thrombin activatable fibrinolysis inhibitor (TAFI) in patients with acute pulmonary embolism. J Thromb Haemost 2003; 1: 492–493.
12. Santamaria A, Martinez-Rubio A, Borrell M, et al. Risk of acute coronary artery disease associated with functional thrombin activatable fibrinolysis inhibitor plasma level. Haematologica 2004; 89:
880–881.
13. Schroeder V, Chatterjee T, Mehta H, et al. Thrombin activatable fibrinolysis inhibitor (TAFI) levels in patients with coronary artery disease investigated by angiography. Thromb Haemost 2002;88:
1020–1025.
14. Juhan-Vague I, Morange PE, Aubert H, et al.; HIFMECH Study Group. Plasma thrombin-activatable fibrinolysis inhibitor antigen concentration and genotype in relation to myocardial infarction in the north and south of Europe. Arterioscler Thromb VascBiol 2002; 22: 867–873.
15. Juhan-Vague I, Morange PE; PRIME Study Group. Very high TAFI antigen levels are associated with a lower risk of hard coronary events: the PRIME Study. J Thromb Haemost 2003;1:2243–2244.
16. Cellai AP, Antonucci E, Alessandrello Liotta A, et al. TAFI activity and antigen plasma levels are not increased in acute coronary artery disease patients admitted to a coronary care unit. Thromb Res 2006; 118: 495–500.
17. Morange PE, Tregouet DA, Frere C, et al.; The Prime Study Group. TAFI gene haplotypes, TAFI plasma levels and future risk of coronary heart disease: the PRIME Study. J Thromb Haemost 2005;3:1503–
1510.
18. Brouwers GJ, Leebeek FW, Tanck MW, et al. Association between thrombin-activatable fibrinolysis inhibitor (TAFI) and clinical outcome in patients with unstable angina pectoris. Thromb Haemost 2003;90: 92–100.
19. Silveira A, Schatteman K, Goossens F, et al. Plasma procarboxy peptidase U in men with symptomatic coronary artery disease. Thromb Haemost 2000;84: 364–368.
20. Montaner J, Ribo M, Monasterio J, et al. Thrombin activable fibrinolysis inhibitor levels in the acute phase of ischemic stroke. Stroke 2003;34: 1038–1040.
21. Santamaria A, Oliver A, Borrell M, et al. Risk of ischemic stroke associated with functional thrombin activatable fibrinolysis inhibitor plasma levels. Stroke 2003; 34: 2387–2391.
22. Leebeek FW, Goor MP, Guimaraes AH, et al. High functional levels of thrombin-activatable fibrinolysis inhibitor are associated with an increased risk of first ischemic stroke. J Thromb Haemost 2005;3: 2211–2218.
23. Ladenvall C, Gils A, Jood K, et al. Thrombin activatable fibrinolysis inhibitor activation peptide shows association with all major subtypes of ischemic stroke and with TAFI gene variation. Arterioscler Thromb Vasc Biol 2007;27: 955–962.
24. Koeleman BP, Reitsma PH, Allaart CF, et al. Activated protein C resistance as an additional risk factor for thrombosis in protein C-deficient families. Blood 1994;84: 1031–1035.
25. Zoller B, Berntsdotter A, Garcia de Frutos P, et al. Resistance to activated protein C as an additional genetic risk factor in hereditary deficiency of protein S. Blood 1995;85: 3518–3523.
26. van Boven HH, Reitsma PH, Rosendaal FR, et al. Factor V Leiden (FVR506Q) in families with inherited antithrombin deficiency. Thromb Haemost1996;75: 417–421.