Title: Variations of hepcidin and iron‐status parameters during the menstrual cycle in healthy women
Abstract: Hepcidin is the central regulator of systemic iron metabolism (Ganz, 2013). In clinical practice, measurements of serum hepcidin (SH) could help to determine the cause of anaemia, iron deficiency and iron overload, to predict iron absorption from food, to optimize the treatment of haemochromatosis and to manage erythropoietin therapy. Interpretation of SH measurement in practice will be dependent on understanding physiological variations of this hormone. Previous studies have suggested that SH varied according to a diurnal cycle (Ganz et al, 2008) and was modified by meals (Schaap et al, 2013). Gender should also be taken into account, as women tend to have lower hepcidin values than men. Moreover, in women, age should be also considered, because values of hepcidin are higher in post- than in pre-menopausal women. Nevertheless, iron status is the main determinant of SH concentration (Galesloot et al, 2011). Blood loss during menses, the main cause of iron deficiency in young women, varies between 20 and 80 ml during a period, representing a loss of 10–40 mg of iron in women with regular menstrual cycles (MC) (Higham et al, 1990). This is significant compared to the unregulated 1 mg eliminated daily through skin, intestinal and urinary cell desquamation, the only other physiological way to eliminate iron. To date, no study has examined whether menses induce significant variations in SH values. The present study (Clinical Trial.gov NCT01764412) was conducted in four French hospitals. Ninety healthy women, aged 18-45 years, with normal iron parameters, regular cycles and normal duration of menses (4 ± 1 days) were included. Fifty-four used oral contraception. Transferrin saturation (TS), serum ferritin (SF), haemoglobin (Hb), serum iron (SI) and SH were measured at six visits distributed throughout the MC. All visits were planned on fasting subjects, between 8am and 9am. Day 0 was defined as the day after menses began. The following 3 visits were scheduled during menses and the following days. The last two visits took place respectively at the middle and at the end of the cycle. Serum hepcidin-25 was quantified using a CE-marked Enzyme Immunoassay (Bachem Inc., Torrance, CA, USA). The intra-assay and inter-assay coefficient of variation was 3·5% and 6·3%, respectively. The lower quantification limit was 0·01 nmol/l. Hepcidin, SI and TS significantly changed during the MC, first dipping during menses, then increasing at mid-cycle before stabilizing during the second part of the cycle (Fig 1). This is in accordance with a previous epidemiological study that excluded women using oral contraceptives, which showed that iron parameters were lowest in women whose blood was drawn during menses (Kim et al, 1993). On the other hand, a small study of 12 women reported no significant differences in iron status across MC (Belza et al, 2005). Our study confirms the results of Kim et al (1993) and extends them to women using oral contraception. Indeed, fluctuations of hepcidin and other iron parameters showed similar patterns whether women used contraceptive pills or not, even if the amplitude of variations were less pronounced in women who used oral contraception. The intensity of blood loss, assessed by the Higham score (Higham et al, 1990), was correlated with the magnitude of hepcidin and iron variations. The higher the score, the larger was the decrease in SI, TS and SH levels during menses and their rebound on subsequent days. The women who used oral contraceptives, known to limit the abundance of menses, had lower Higham scores than non-users. This could also explain that the amplitude of variations of iron parameters during MC was less pronounced in these women. In order to provide recommendations on the timing measurements of SH and iron during the menstrual cycle, the data were modelled using non-linear mixed effect models (Kuhn & Lavielle, 2005). A joint model, describing the simultaneous evolution of iron and hepcidin during MC was developed, and showed that both molecules conformed to a turnover model with time-varying secretion and elimination rate constants. The simulated profiles of iron and hepcidin concentrations were used to compute the percentage of variation of these two molecules with respect to the value at the end of the cycle over different time periods. The measurement at the end of the cycle (last visit scheduled between Day 26 and Day 29) was considered to be the reference value as concentrations were stable during this period. Figure 2 shows that fluctuations were large during menses and during the second fortnight of MC, with 40% of women experiencing at least 30% variations compared to baseline. On the other hand, during the last five days of MC less than 30% variations were observed for all simulated profiles. Our observations could have practical implications in both daily practice and research. For example, the diagnosis and screening of genetic haemochromatosis, which are usually based on the determination of C282Y mutation in the HFE gene in subjects with TS >45%, should take into account our results. Indeed, on the days following menses corresponding to the peak of hepcidin, TS and iron, TS higher than 45% was observed in more than 13% of women compared to only 1% during menses and at the end of the cycle. Therefore measuring TS during the days following menses could result in an excess of HFE genotyping. The diagnosis of iron deficiency should also take fluctuations of iron parameters during the MC into account. We observed that 29% of the participants had iron levels that were less than normal during menses. Finally, fluctuations of hepcidin and iron parameters during the MC should also be taken into account when these biological variables serve as judgement criteria in clinical trials and epidemiological studies recruiting women. It is of note that the hepcidin:SF ratio remained fairly stable during the cycle. In conclusion, we suggest that hepcidin and iron-status parameters should not be measured during menses or on the following days, but during the last five days of the menstrual cycle. Moreover, hepcidin should not be interpreted in isolation from other measures of iron status. This work was supported by grants from the Projet Inter-régional Hospitalier de Recherche Clinique 2013. We thank Pr Jean Levêque for his help in study conception and the Centre de Ressources Biologiques of Rennes for its support in the processing of biological samples. Fabrice Lainé contributed to the study design, study execution, data interpretation, and wrote the manuscript. Adeline Angeli performed statistical analyses and contributed to writing the manuscript. Martine Ropert performed biochemical analyzes. Bruno Laviolle contributed to the methodology. Caroline Jezequel, Edouard Bardou-Jacquet, Yves Deugnier contributed to data interpretation, and wrote the manuscript. Sylvie Sacher-Huvelin, Karine Lacut and Valérie Gissot contributed to the study execution. Audrey Lavenu contributed to data analyses. Emmanuelle Comets contributed to the study design, data analyses and interpretation, and wrote the manuscript. The authors declare no conflict of interest.