In CHD patients, it is usually difficult to control anemia because of a complex derangement of iron metabolism, which is due to chronic inflammation, blood losses, and concomitant Epo administration
. Increased serum levels of hepcidin, the hepatic hormone regulating iron metabolism, have been suggested to contribute to the functional iron deficiency that limit erythropoiesis in CHD
[2, 3, 31]. Differently from what reported in smaller series of patients with unmatched controls, including previous studies from our group
[5–8, 32, 33], we found that hepcidin-25 levels were not significantly increased in the whole CHD population. Nevertheless, accordingly to the previous reports
[5–8, 32, 33], the major determinants of hepcidin were serum ferritin and CRP levels. The failure to confirm a relative hyper-hepcidinemia in the overall CHD cohort could be explained by a number of reasons: i) the inclusion of a large and less selected CHD population compared to previous studies, more closely reflecting patients observed in clinical practice, including those requiring high doses of Epo and with relative iron deficiency (factors both known to reduce hepcidin), as suggested by higher hepcidin levels in patients without functional iron deficiency; ii) the relatively low average iron stores in this cohort (as reflected by median serum ferritin levels of only 265 ng/ml, Table
1) because of a local policy aimed at minimizing iron supplementation, due to long standing interest in iron metabolism and the side effects of iron overload; and iii) at variance with previous studies, e.g. Zaritsky et al.
, the meticulous matching of the controls for age and gender, recently established as major determinants of hepcidin-25 at population level
[20, 34], as well as the systematic exclusion of even subclinical iron deficiency in controls, which both contributed to a more realistic comparison of hepcidin-25 levels than those made until now. Accordingly with these considerations, however, when CHD patients with relative iron deficiency were excluded from these analyses, hepcidin-25 was actually higher in CHD than in controls. Anyway, the comparison of hepcidin levels between CHD patients and controls was not the main aim of the present study, which was not therefore specifically designed to achieve this goal. Our results in this sense need to be confirmed in similarly large patient populations and matched controls.
Clearance of hepcidin by hemodialysis
 may possibly compensate for increased production, and explain the reduced H/F ratio in patients compared to controls. Notwithstanding, in patients without severe iron deficiency and active inflammation at the time of evaluation, hepcidin was associated with lower MCV, i.e. with iron-restricted erythropoiesis, suggesting that it negatively influences iron availability to the erythron, and that it represents a potential therapeutic target to improve anemia management.
The specific aim of this study was to evaluate whether the A736V TMPRSS6 polymorphism regulating hepcidin transcription, a determinant of iron-restricted erythropoiesis in the general population
[21, 35], influences hepcidin levels and erythropoiesis in CHD. To increase the power of this analysis, patients were stratified for the presence of loss-of-function HFE mutations, that we preliminarily confirmed to influence hepcidin in this series
. The major finding was that the 736 V TMPRSS6 loss-of-function variant appears to modulate the effect of HFE mutations on hepcidin. Indeed, the A736V polymorphism influenced serum hepcidin in patients positive for HFE mutations. This suggests that the 736 V variant with defective proteolytic activity determining increased hepcidin transcription
 may abrogate the inhibitory effect of HFE mutations on hepcidin. Thus, the A736V TMPRSS6 variant appears as a modifier of the phenotypic expression of HFE mutations in patients with CHD, who are characterized by chronic subclinical inflammation.
Furthermore, in patients without overt iron deficiency and acute inflammation, the 736 V variant was associated with higher hepcidin levels and with higher requirement of Epo for anemia management, thus suggesting that the effect of TMPRSS6 genotype translates into clinically detectable differences in erythropoiesis. Importantly, at multivariate analysis the association between TMPRSS6 genotype and Epo maintenance dose was independent of subclinical inflammation, as indicated by CRP levels. These data are in line with the association between TMPRSS6 736 V with hepcidin levels, and in turn with the positive association of hepcidin with the Epo maintenance dose in the same subgroup. Therefore, inhibition of hepcidin might be helpful for a better control of anemia in patients predisposed to high hepcidin release
. Evaluation of the impact of HFE and TMPRSS6 genotype on the survival of CHD patients after adequate follow-up would be instrumental to fully define their clinical impact.