SEARCH

SEARCH BY CITATION

Keywords:

  • all-cause mortality;
  • factor XII;
  • overall survival;
  • risk prediction;
  • U-shaped association

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Disclosure of Conflict of Interests
  9. References

Summary. Introduction: The clinical relevance of decreased coagulation factor XII (FXII) plasma activity as a risk factor for both venous and arterial thrombosis is still discussed controversially. The current study evaluated the predictive value of FXII levels for all-cause mortality in a large Viennese patient cohort. Patients and methods: Individuals, whose FXII activity levels were determined for suspected coagulation disorders or thrombophilia screening between 1991–2003 were included in this study (n = 8936, 51% male, 49% female, median age 43 years). Death/survival was determined by record linkage with the Austrian Death Registry. The median observation period was 5 years covering a total of 46 400 person years; the death rate was 17.1%. For Cox regression analysis, FXII plasma activity was divided into 11 categories of 10% steps with the category of > 100% FXII serving as a reference category. Results: With decreasing FXII plasma activity, hazard ratios for all-cause mortality gradually increased linearly from 1.0 in the > 100% category to 1.5 (95% CI: 1.2–1.9) in the 80–90% category to 4.7 (95% CI: 3.4–6.5) in the 10–20% category. Similar results were obtained, when only vascular mortality or death as a result of ischemic heart disease was considered. No significant increase in all-cause mortality (HR: 1.4, 95%CI 0.7–2.8) was observed in the small group of FXII-deficient subjects [0–10% category (= 58)]. Conclusions: This study first demonstrates a strong and almost linear association of FXII plasma activity between 90% and 10% with all-cause mortality in a large Viennese patient cohort. Interestingly, mortality rates are not increased when FXII activity is below 10%, resulting in a U-shaped survival curve.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Disclosure of Conflict of Interests
  9. References

Coagulation factor XII (FXII; Hageman factor) is an 80-kDa serine protease with an average plasma concentration of 30 μg mL−1. Contact of FXII with negatively charged surfaces leads to proteolytic cleavage and activation of the FXII molecule. The generated FXIIa can initiate activation of FXI and also seems to participate in the conversion of plasminogen to plasmin [1,2]. While, FXII plays a central role in the initiation of coagulation and fibrinolysis in vitro, the physiological function of FXII is still under discussion. Severe deficiency of FXII does not lead to an increased bleeding tendency, and in vivo FXII does not seem to be essential for coagulation activation. Rather, as a result of its role in fibrinolysis low FXII levels or genetic polymorphisms leading to lower FXII plasma activity [3,4] have been associated with an increased risk for venous [5–8] as well as for arterial thromboembolism [9–12] in some studies. However, this could not be confirmed in other studies, in which no effect of FXII levels on thrombosis was observed [13–18]. Thus, the role of decreased FXII levels as a potential predictor of arterial or venous thrombosis remains unclear.

Owing to conflicting data, the role of subnormal and low FXII levels as potential risk factors for arterial and venous thrombosis has been a matter of intense debate as several studies yielded rather controversial results. This could be the result of limited sample size of the studies as well as variations of the chosen endpoints ranging from superficial phlebitis to ischemic stroke. Until now, the impact of FXII levels on all cause or cardiovascular mortality has not been analyzed, although this would allow more accurate risk estimation, as a result of an objectively defined endpoint. We evaluated the relevance of FXII plasma activity as a potential predictor for all-cause mortality in a large-scale Austrian cohort.

Aim of the study

Because decreased FXII levels have been associated with impaired fibrinolysis and an increased risk of thrombosis, we hypothesized that lower FXII plasma activity might predict long-term all-cause mortality. To evaluate its impact upon cardiovascular events, we included overall vascular and cardiovascular mortality as secondary endpoints in our study.

Patients and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Disclosure of Conflict of Interests
  9. References

Patients

All first ever admissions to the Clinical Institute of Medical and Chemical Laboratory Diagnostics for determination of coagulation FXII activity between January 1991 and July 2003 were included in our study. FXII determination in most patients was performed for reasons of prolonged partial activated thromboplastintime (60% of all patients presented with prolonged APTT), thrombophilia screening or a suspected coagulation disorder upon the clinicians decision. Inclusion criteria were a valid FXII plasma activity test and complete patient data including gender, name and date of birth required for successful record linkage. Exclusion criteria were incomplete patient data (patient’s name, date of birth or valid FXII levels) and admission from an intensive care unit. Record linkage was performed via database query of the Austrian death registry, resulting in date of death (if occurred between January 1990 and December 2004) and cause of death encoded either according to the International Code of diseases Version 9 (ICD9) before 2002, later on according to the International Code of diseases Version 10. The Austrian death registry comprises all deaths within Austria and the deaths of Austrian citizens in foreign countries if reported to Austrian officials. According to Austrian laws all deaths have to undergo autopsy, if the final cause of death is not evident from the patients’ history resulting in an overall post-mortem frequency of 51% in our study. For statistical analysis, only anonymized data were used containing no personal information except age in years and gender. The study was approved by the ethics committee of the Medical University of Vienna.

Coagulation tests

Citrated platelet poor plasma was obtained by atraumatic venupuncture.

Factor XII clotting activities were determined by an one-step clotting assay using FXII-deficient plasma (Dade Behring, Marburg, Germany) and the APTT reagent Dade® Actin®-FS (Dade Behring) on a KC-10 coagulometer (Amelung; Lemgo-1, Lieme, Germany) as described previously [4].

The assay was standardized by comparing the results with pooled reference plasma samples retrieved from healthy controls. Together with each batch of patient samples tested the respective normal and low plasma controls (Preciclot I & II; Roche Diagnostics, Basel, Switzerland) were analyzed. For calibration, the STA Unicalibrator (Diagnostica Stago, Asnieres, France) was used. The observed interassay variation was 5–15%. During the time course of the study, the assay remained essentially unchanged except for standard human plasma (Dade Behring Inc., Deerfield, IL, USA) which has been used for calibration until 1996.

Prothrombin time (PT), fibrinogen, total cholesterol, triglycerides and C-reactive protein (CRP) were evaluated using the respective reagents [Dade Behring for PT, all others from Roche Diagnostics (Roche Diagnostics, Mannheim, Germany)].

Determination of outcome variables and statistical analysis

The main outcome variable was all-cause mortality, defined as death occurring after FXII determination and before 31 December 2004, regardless of diagnosis. Non-cancer mortality was defined as death occurring for other causes than neoplasia (defined as ICD9 groups 140–239 and ICD10 groups C00 to D99); all-cause vascular mortality was considered present in case of ICD9 codes 390–459 respectively ICD10 group I00 to I99. Mortality as a result of ischemic heart disease was defined as ICD9 groups 410–416 and ICD10 groups I20–I25. Death as a result of cerebrovascular disease was defined as ICD9 groups 430–438 and ICD10 group I60–I69. Observation time was calculated in years from FXII determination to death or until the end of the observation time (31 December 2004) in case of survivors. Age was calculated at time of FXII determination.

To facilitate analysis, FXII activity levels were categorized into categories for each 10% increase in plasma activity. Individuals with FXII activity above 100% were grouped together and served as a reference category. The rationale for this categorization is the previously reported association of FXII deficiency with venous or arterial thrombosis. Based on these results, individuals with FXII levels of 100% or higher should have the lowest risk for suffering from thrombosis. The influence of FXII levels on our main outcome variable all-cause mortality was assessed in a multivariate Cox regression adjusting for gender and age as possible confounders.

In a second step, we evaluated the influence of possible confounders (PT as a marker of potential general decrease of coagulation markers in liver disease, fibrinogen and CRP, cholesterol, triglycerides and diagnosis of diabetes, hypertension, coronary artery disease or neoplasia at time of FXII determination, where data available). To facilitate multivariate analysis, continuous variables (PT, fibrinogen, CRP and plasma lipids) were categorized in quintiles with the lowest category serving as a reference category except for PT, where the highest category was set as reference category. Additionally, missing values were assigned to a separate category to allow multivariate analysis.

Regression diagnostics were performed according to standard recommendations. Correlations were calculated using Spearman’s rank sum correlation coefficient and univariate comparisons were made using the Mann–Whitney U-test or chi-square test where appropriate. A two-sided P-value < 0.05 was considered statistically significant. Unless otherwise stated, all continuous variables are given as median and interquartile ranges (IQR) and categorical variables as counts and percentages. All calculations were performed using spss 12.0 software (SPSS Inc., Chicago, IL, USA).

Estimate of sample bias

To give an estimate of a possible sample bias of our hospital-based sample, we compared the observed overall survival in our study normalized for estimated survival in the general Austrian population. For this purpose, we calculated the individual probability of death within the hypothetical maximum observation period (date of analysis until the last day included in the record linkage analysis) considering individual age, gender and maximum observation time using the official Austrian mortality tables provided by Statistik Austria (http://www.statistik.at/fachbereich_03/Stt2000_2002.xls, 2005). Normalized overall survival was defined as 1 (all-cause mortality – estimated mortality) for each individual and mean normalized overall survival and its respective 99% confidence intervals were calculated and plotted in Fig. 1.

image

Figure 1.  Overall survival within the factor (F) XII categories, normalized to the Austrian population. Values near 1 indicate that overall survival was near those observed within the Austrian population, values < 1 indicate a decreased survival compared with the general population. While no larger differences in survival were observed in patients with high FXII levels, the gap increased with decreasing FXII levels. Notably no significant differences were observed in individuals with severe FXII deficiency (< 10%).

Download figure to PowerPoint

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Disclosure of Conflict of Interests
  9. References

From a total of 9433 patients in whom FXII levels have been determined, 8936 patients could be included in our study. Median age was 43 years (ranging from 0 to 96 years) at time of analysis and the median observation period was 5 years resulting in 46 400 person years (Table 1). A total of 1526 deaths were recorded. Factor XII levels varied with a broad inter-individual range with a median of 78% (IQR: 53–108%; mean FXII levels: 80%, standard deviation: ± 35%). Severe FXII deficiency with FXII levels ≤ 10% was seen in 58 patients. In a multivariate Cox regression adjusted for gender and age as possible confounders, FXII levels were significantly associated with all-cause mortality (Fig. 2). Compared with individuals within the reference category (FXII > 100%), hazard ratios gradually increased from 1.5 (95% CI 1.3–1.9) in the 80–90% category to 4.7 (% CI 3.4–6.6) in the 10–20% category (Figs 1 and 2). Interestingly, in individuals with very low FXII levels (< 10%, lowest category) all-cause mortality was comparable to the reference category (Fig. 3). The association between FXII levels and mortality remained essentially unchanged when considering non-cancer mortality, all-cause vascular mortality or mortality as a result of ischemic heart disease (and with a limited number of cases as a result of venous thromboembolism) as secondary endpoints (Fig. 1, Supplementary table). Even after adjustment for other plasmatic factors (PT, fibrinogen, cholesterol, triglycerids) or clinical diagnoses such as hypertension and diabetes these results did not change significantly (Supplementary table).

Table 1.   Patients’ characteristics of the study population
 Patients characteristics (n = 8936)
  1. IQR, interquartile ranges.

Gender (male/female)4556/4380 (51%/49%)
Median age in years (IQR, range)43 (IQR: 22–61)
Median FXII plasma activity (IQR)78 (IQR: 53–108)
Median observation period in years (IQR)4.6 (IQR: 2.6–8.0)
All-cause mortality (cases, %)1526 (17%)
Non-cancer mortality (cases, %)922 (10%)
All-cause vascular mortality (cases, %)407 (5%)
Mortality as a result of ischemic heart disease (cases, %)159 (2%)
FXII plasma activity categories
> 100%2451 (27%)
> 90% ≤ 100%819 (9%)
> 80% ≤ 90%961 (11%)
> 70% ≤ 80%942 (11%)
> 60% ≤ 70%928 (10%)
> 50% ≤ 60%811 (9%)
> 40% ≤ 50%750 (8%)
> 30% ≤ 40%667 (8%)
> 20% ≤ 30%404 (5%)
> 10% ≤ 20%145 (2%)
≤ 10%58 (< 1%)
image

Figure 2.  Factor (F) XII levels and all-cause mortality. Survival and all-cause mortality in a Cox regression adjusted for gender and age. All-cause mortality increased progressively with decreasing factor XII (FXII) levels (indicated right, for numbers at risk please refer to Table 1). Within the group with lowest FXII (≤ 10%) survival rates were comparable to individuals within the group with highest FXII levels.

Download figure to PowerPoint

image

Figure 3.  Hazard ratios for different endpoints (95% CI) were obtained from Cox regression analysis after adjustment for age and gender as possible confounders. In all chosen endpoints [all-cause mortality, non-cancer mortality, all-cause vascular mortality (including both the venous and arterial system), and death as a result of ischemic heart disease], hazard ratios for unfavorable outcome gradually increased with decreasing Factor (F) XII levels. Notably, in individuals within the lowest category (< 10%) mortality reverted to almost normal.

Download figure to PowerPoint

As described previously [19], FXII plasma activity correlated positively with total cholesterol (R = 0.43), PTZ (R = 0.15) and negatively with CRP levels (R = −0.29, all P < 0.001), whereas all other variables (gender, age, triglycerides, fibrinogen, diagnosis of diabetes, hypertension or coronary artery disease) only showed weak (R < 0.1) or no significant correlation with FXII plasma activity.

Within the reference category (FXII > 100%) no significant differences in overall mortality were observed after stratification according to the FXII level (data not shown).

To provide an estimate for sampling bias in our analysis, we analyzed normalized overall survival according to FXII categories (Fig. 2). Within the reference group there were no major differences between both values whereas in the lower FXII categories the survival in comparison to the Austrian population decreased in parallel to increasing hazard ratios in the Cox regression analysis. Again, survival in patients with severe FXII deficiency (< 10%) was not different from the reference category. To detect a potential interaction of FXII activity with observation time patients were stratified according to their observation time in years. Hazard ratios for moderate FXII deficiency (defined as 10–50% activity) remained essentially unchanged from 1.6 (95% CI: 1.3–2.0) for fatal outcome within the first year, to 1.7 (95% CI 1.1–2.7) for death after the fourth year post-FXII determination after adjustment for gender and age (Fig. 3).

The association of FXII category and all-cause mortality remained essentially unchanged after stratification according to gender or age (divided in quartiles, data not shown).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Disclosure of Conflict of Interests
  9. References

The biological function of coagulation FXII is poorly understood. Divergent biological functions of FXII have been described and controversial results regarding the clinical consequences of low FXII activity have been reported. While lower FXII levels have been associated with a higher risk for arterial or venous thrombosis [7,10–12,14,20,21], no significant increase in thrombotic risk could be observed in individuals with severe FXII deficiency [13]. Data from our study are in line with this paradoxon. Whereas individuals with FXII activity > 100% showed an overall survival comparable to the Austrian population (Fig. 1), hazard-ratios for death gradually increased with decreasing FXII activity (Fig. 3, Supplementary table). Individuals with FXII activities 10–30% had a more than 4-fold risk of dying compared with the reference category. This association remained essentially unchanged when considering vascular death or death as a result of ischemic heart disease as primary outcome variables or after adjustment for other plasmatic factors or clinical diagnoses.

However, in individuals with severe FXII deficiency (< 10%) the hazard ratio for death did not differ significantly from the reference category (> 100%). Although this patient category comprises relatively few patients (n = 58), the number of individuals should be sufficient to detect an at least 3- to 5-fold increase in mortality.

We can only speculate on the reasons for these findings. Conceivably, there is a parallel decrease in fibrinolytic and procoagulant activity in individuals with severe FXII deficiency. In individuals with FXII levels above 10%, increased procoagulant activity is accompanied by a concomitant increase in thrombotic potential. This has also been demonstrated in a recent study by Rennéet al. [22,23] in a mouse model. Knock-out mice lacking FXII showed no increased bleeding tendency, while intravascular thrombus formation was significantly decreased. Mice with FXII activity of ∼50% formed normal thrombi upon challenge with prothrombotic measures. The thrombotic potential increased with FXII levels.

Strengths and limitations of this study

To our knowledge this is the first study which evaluated the association between FXII and mortality in a large population. The chosen record linkage approach offers the unique opportunity to evaluate the outcome of risk markers in a large population (n = 8936) over a long period of time yielding an observation period of 46 400 person years. As Austrian laws require that all deaths have to be recorded in the central death registry, this approach allows an almost complete follow-up of all patients. The only losses might occur as a result of spelling errors in names, which result in faulty record linkage, or missing persons, who are not recorded as deaths until 50 years after the date of disappearance. Overall, we estimate that these losses affect less than 1% of the study population and are negligible for statistical analysis. In contrast to clinical diagnoses which are subject to examiner bias and usually vary because of different diagnostic criteria, death is usually reliably recorded and misdiagnoses rarely occur. Owing to the high post-mortem frequency (51%) in our patient group, and the Austrian legal situation, we estimate that diagnoses leading to death are recorded correctly in ∼95% of the cases.

Nevertheless, several limitations apply to our study. Up to date, FXII levels have not been assayed in large unbiased population based cohorts, thus, the range of variation of FXII levels in the Middle European population is unknown. In our patient group 23% had FXII plasma activity < 50%. Although most of these cases are probably caused by a frequent C46T variant in the 5′ untranslated region of the FXII gene [4], we cannot exclude that other factors such as strong lupus anticoagulant, specific FXII antibodies or end-stage liver disease also contribute to this finding. Yet, as these diseases are rather infrequent, a major bias in this group seems unlikely.

As we only considered a single measurement of FXII activity we are unable to distinguish between acquired and inherited FXII deficiency. Thus, the association of FXII levels with mortality does not necessarily imply a causal relationship, which would require a study based on Mendelian randomization. However, this is beyond the scope of our study, as we aimed to evaluate whether FXII levels are risk markers for all-cause mortality.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Disclosure of Conflict of Interests
  9. References

Data from this large study showed that FXII plasma activity is strongly associated with all-cause mortality. In comparison to individuals presenting with FXII levels above 100%, hazard ratios for death continuously increased to almost 5-fold in individuals with levels between 10% and 20%. Interestingly, in individuals with severe FXII deficiency defined as FXII < 10% mortality was comparable to the reference category and to the Austrian population.

Disclosure of Conflict of Interests

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Disclosure of Conflict of Interests
  9. References

The authors state that they have no conflict of interest.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Disclosure of Conflict of Interests
  9. References
  • 1
    Braat EA, Dooijewaard G, Rijken DC. Fibrinolytic properties of activated FXII. Eur J Biochem 1999; 263: 90411.
  • 2
    Schousboe I, Feddersen K, Rojkjaer R. Factor XIIa is a kinetically favorable plasminogen activator. Thromb Haemost 1999; 82: 10416.
  • 3
    Kanaji T, Okamura T, Osaki K, Kuroiwa M, Shimoda K, Hamasaki N, Niho Y. A common genetic polymorphism (46 C to T substitution) in the 5’-untranslated region of the coagulation factor XII gene is associated with low translation efficiency and decrease in plasma factor XII level. Blood 1998; 91: 20104.
  • 4
    Endler G, Exner M, Mannhalter C, Meier S, Ruzicka K, Handler S, Panzer S, Wagner O, Quehenberger P. A common C– > T polymorphism at nt 46 in the promoter region of coagulation factor XII is associated with decreased factor XII activity. Thromb Res 2001; 101: 25560.
  • 5
    Gallimore MJ, Harris SL, Jones DW, Winter M. Plasma levels of factor XII, prekallikrein and high molecular weight kininogen in normal blood donors and patients having suffered venous thrombosis. Thromb Res 2004; 114: 916.
  • 6
    Lammle B, Wuillemin WA, Huber I, Krauskopf M, Zurcher C, Pflugshaupt R, Furlan M. Thromboembolism and bleeding tendency in congenital factor XII deficiency – a study on 74 subjects from 14 Swiss families. Thromb Haemost 1991; 65: 11721.
  • 7
    Soria JM, Almasy L, Souto JC, Bacq D, Buil A, Faure A, Martinez-Marchan E, Mateo J, Borrell M, Stone W, Lathrop M, Fontcuberta J, Blangero J. A quantitative-trait locus in the human factor XII gene influences both plasma factor XII levels and susceptibility to thrombotic disease. Am J Hum Genet 2002; 70: 56774.
  • 8
    Tirado I, Soria JM, Mateo J, Oliver A, Souto JC, Santamaria A, Felices R, Borrell M, Fontcuberta J. Association after linkage analysis indicates that homozygosity for the 46C– > T polymorphism in the F12 gene is a genetic risk factor for venous thrombosis. Thromb Haemost 2004; 91: 899904.
  • 9
    Doggen CJ, Rosendaal FR, Meijers JC. Levels of intrinsic coagulation factors and the risk of myocardial infarction among men: opposite and synergistic effects of factors XI and XII. Blood 2006; 108: 404551.
  • 10
    Halbmayer WM, Haushofer A, Radek J, Schon R, Deutsch M, Fischer M. Prevalence of factor XII (Hageman factor) deficiency among 426 patients with coronary heart disease awaiting cardiac surgery. Coron Artery Dis 1994; 5: 4514.
  • 11
    Zito F, Lowe GD, Rumley A, McMahon AD, Humphries SE. Association of the factor XII 46C > T polymorphism with risk of coronary heart disease (CHD) in the WOSCOPS study. Atherosclerosis 2002; 165: 1538.
  • 12
    Santamaria A, Mateo J, Tirado I, Oliver A, Belvis R, Marti-Fabregas J, Felices R, Soria JM, Souto JC, Fontcuberta J. Homozygosity of the T allele of the 46 C- > T polymorphism in the F12 gene is a risk factor for ischemic stroke in the Spanish population. Stroke 2004; 35: 17959.
  • 13
    Zeerleder S, Schloesser M, Redondo M, Wuillemin WA, Engel W, Furlan M, Lammle B. Reevaluation of the incidence of thromboembolic complications in congenital factor XII deficiency – a study on 73 subjects from 14 Swiss families. Thromb Haemost 1999; 82: 12406.
  • 14
    Von Kanel R, Wuillemin WA, Furlan M, Lammle B. Factor XII clotting activity and antigen levels in patients with thromboembolic disease. Blood Coagul Fibrinolysis 1992; 3: 55561.
  • 15
    Endler G, Mannhalter C, Sunder-Plassmann H, Lalouschek W, Kapiotis S, Exner M, Jordanova N, Meier S, Kunze F, Wagner O, Huber K. Homozygosity for the C– > T polymorphism at nucleotide 46 in the 5’ untranslated region of the factor XII gene protects from development of acute coronary syndrome. Br J Haematol 2001; 115: 10079.
  • 16
    Colhoun HM, Zito F, Norman Chan N, Rubens MB, Fuller JH, Humphries SE. Activated factor XII levels and factor XII 46C > T genotype in relation to coronary artery calcification in patients with type 1 diabetes and healthy subjects. Atherosclerosis 2002; 163: 3639.
  • 17
    Kohler HP, Futers TS, Grant PJ. FXII (46C– > T) polymorphism and in vivo generation of FXII activity – gene frequencies and relationship in patients with coronary artery disease. Thromb Haemost 1999; 81: 7457.
  • 18
    Oguchi S, Ito D, Murata M, Yoshida T, Tanahashi N, Fukuuchi Y, Ikeda Y, Watanabe K. Genotype distribution of the 46C/T polymorphism of coagulation factor XII in the Japanese population: absence of its association with ischemic cerebrovascular disease. Thromb Haemost 2000; 83: 1789.
  • 19
    Zito F, Drummond F, Bujac SR, Esnouf MP, Morrissey JH, Humphries SE, Miller GJ. Epidemiological and genetic associations of activated factor XII concentration with factor VII activity, fibrinopeptide A concentration, and risk of coronary heart disease in men. Circulation 2000; 102: 205862.
  • 20
    Castaman G, Ruggeri M, Tosetto A, Missiaglia E, Rodeghiero F. Thrombosis in patients with heterozygous and homozygous factor XII deficiency is not explained by the associated presence of factor V Leiden. Thromb Haemost 1996; 76: 275.
  • 21
    Halbmayer WM, Haushofer A, Schon R, Mannhalter C, Strohmer E, Baumgarten K, Fischer M. The prevalence of moderate and severe FXII (Hageman factor) deficiency among the normal population: evaluation of the incidence of FXII deficiency among 300 healthy blood donors. Thromb Haemost 1994; 71: 6872.
  • 22
    Pauer HU, Renne T, Hemmerlein B, Legler T, Fritzlar S, Adham I, Muller-Esterl W, Emons G, Sancken U, Engel W, Burfeind P. Targeted deletion of murine coagulation factor XII gene-a model for contact phase activation in vivo. Thromb Haemost 2004; 92: 5038.
  • 23
    Renne T, Pozgajova M, Gruner S, Schuh K, Pauer HU, Burfeind P, Gailani D, Nieswandt B. Defective thrombus formation in mice lacking coagulation factor XII. J Exp Med 2005; 202: 27181.