To cite this article: Kanaji T. Lower factor XII activity is a risk marker rather than a risk factor for cardiovascular disease: a rebuttal. J Thromb Haemost 2008; 6: 1053-4. See also Bach J, Endler G, Mannhalter C, Hellstern P. Coagulation factor XII activity, activated factor XII, distribution of factor XII C46T gene polymorphism and coronary risk: reply to a rebuttal. This issue, pp 1055-6; Bach J, Endler G, Winkelmann BR, Boehm BO, Maerz W, Mannhalter C, Hellstern P. Coagulation factor XII (FXII) activity, activated FXII, distribution of FXII C46T gene polymorphism and coronary risk. J Thromb Haemost 2008; 6: 291-6. In a recent paper published by this journal, Drs Bach and Endler conducted a large prospective surveillance study to examine the relationships between plasma levels of factor (F) XII activity, activated FXII (FXIIa), the 46C>T ()4c>t) polymorphism in the FXII gene, and coronary heart disease (CHD) [1]. They observed that patients with CHD had significantly lower FXII activity than the controls and concluded that lower FXII activity is an independent risk factor for CHD. Doggen et al. [2] have also reported that patients with myocardial infarction have lower FXII activity than control subjects. They speculated that lower FXII activity may enhance thrombus formation by decreasing fibrinolysis. In contrast, we speculate that reduced FXII activity is not the cause of thrombosis, but the result of it. Our reasoning is as follows. We previously showed the existence of the FXII sequence variant 46C>T ()4c>t), where the T allele creates a novel methionine initiating codon that reduces the translation efficiency of FXII [3]. As a result, the FXII 46C>T polymor-phism is not merely a marker, rather it is a strong determinant of plasma FXII levels. Notably, despite the importance of FXII genotype on FXII levels, the studies by Drs Bach and Endler and others did not observe an association between the FXII 46T/T genotype and CHD. This would suggest that lower FXII activity is not a risk factor, rather it simply represents a risk marker. Supporting this notion is that we performed a case-control study involving deep vein thrombosis (DVT) patients and normal controls to examine the effect of the 46C>T genotype on plasma FXII activity and antigen [4]. We first observed that there was a 46T ()4 t) allele-specific dose-dependent effect on plasma FXII activity and antigen levels, with the T/T patients and controls having lower levels than C/C patients and controls, respectively, while the C/T patients and controls had intermediate levels (Tables 1A and 1B) as we have previously shown [3]. We further found that the patients of the C/C, C/T, and T/T genotypes had lower FXII activity and antigen levels than the controls with the same genotypes; this was statistically significant for the T/T and/or C/T subjects (Tables 1A and 1B). None the less, carriers of 46C/C ()4C/C), indicating a high level of FXII, had an increased risk of DVT (odds ratio 46C/C carriers 2.69, 46T/T as reference) [4]. We also obtained seemingly con...