Reply

Abstract

without iron overload (Fig. 1). (2) As expected, we found an inverse correlation between intestinal iron absorption and plasma hepcidin (r 5 20.61, P < 0.001), ultrasensitive C-reactive protein levels (r 5 20.52, P < 0.001), and serum ferritin (r 5 20.51, P < 0.001). These results indicate that regulation of iron absorption through hepcidin is preserved in DIOS.The study of Hoki et al.(1) raises several issues: (1) the time elapsed between the oral absorption test and the duodenal biopsy was not indicated (a long time could explain the lack of correlation between the oral iron absorption test, serum hepcidin, and expression of ferroportin); (2) the authors did not compare results of the iron absorption test according to the absence or presence of hepatic iron accumulation; (3) the expression of DMT1 in duodenal cells depending on local hypoxia through induction of hypoxia inducible factor2a, (3) the hypothesis of a serum factor regulating DMT1 expression needs further investigation; and (4) the results are difficult to reconcile with the demonstration that a 6-week regimen of a high-fat/highenergy diet (4) induced hepcidin down-regulation in an animal model.It is likely that iron metabolism varies along the long story of the metabolic syndrome: in the constitutive phase of overweight, hepcidin down-regulation allows iron accumulation, probably to supply the increase of erythropoiesis, as has been demonstrated in rats (4) ; and in the steady overweight phase, hepcidin is upregulated, which is related to iron excess. Interpretation of future studies of iron metabolism in DIOS or/and NASH should take such a dynamic view into account. REPLY:The study reported by Ruivard et al.(1) differed markedly from our own, (2) the major difference being the study subjects. Patients with dysmetabolic iron FIG. 1. Intestinal iron absorption evaluated by the area under the curve of 58 Fe given orally in patients with insulin-resistance iron overload (DIOS), overweight controls, and lean controls. Results are expressed in mean 6 standard deviation. Differences between DIOS patients and controls were significant (P < 0.01 for DIOS versus lean controls and P < 0.05 for DIOS versus overweight controls). Abbreviations: AUC, area under the curve; L, lean controls; OW, overweight controls. (Adapted from Ruivard et al.,