Prolonged Chronic Stress and Persistent Iron Dysregulation Prevent Anemia Recovery Following Trauma

Abstract

Introduction

Following major trauma, persistent injury-associated anemia is associated with organ failure, increased length of stay and mortality. We hypothesize that prolonged adrenergic stimulation following trauma is directly responsible for persistent iron dysfunction that impairs anemia recovery.

Materials and Methods

Naïve rodents, lung contusion and hemorrhagic shock followed by daily handling for 13 d (LCHS), LCHS followed by 6 d of restraint stress and 7 d of daily handling (LCHS/CS-7) and LCHS/CS followed by 13 d of restraint stress with day and/or night disruption (LCHS/CS-14) were sacrificed on day 14. Hemoglobin, plasma, urine, bone marrow/liver inflammatory and erythropoietic markers were analyzed.

Results

LCHS/CS-14 led to a significant decline in weight gain and persistently elevated plasma and urine inflammatory markers. Liver IL-6, IL-1β and hepcidin expression were significantly increased following LCHS/CS-14. LCHS/CS-14 also had impaired anemia recovery with reduced plasma transferrin and erythropoietin receptor expression.

Conclusion

Prolonged chronic stress following trauma/hemorrhagic shock led to sustained inflammation with increased expression of IL-1β, IL-6 and hepcidin with decreased iron availability for uptake into erythroid progenitor cells and a lack of anemia recovery.

Introduction

Patients suffering from major trauma often develop a persistent injury-associated anemia, which can last up to several months in severe cases.^1,2 This leads to an increase in blood transfusions, which is associated with higher degrees of organ failure, lengths of stay, as well as mortality.^3-5 These critically ill trauma patients who stay in the intensive care unit remain hyperadrenergic due to the severity of injury, repeated operations, mechanical ventilation, invasive procedures, severe pain, and the loss of traditional sleep wake patterns^6,7. While the initial sympathetic response to injury is physiologic, when the magnitude and duration of this catecholamine stimulation is prolonged, the response becomes pathologic⁸. Prior research has demonstrated that this persistent-injury associated anemia is driven by a prolonged inflammatory response, which is mediated by a norepinephrine-driven hypercatecholamine state.^9-13 This clinical condition has been ‘reverse-translated’ to a rodent model of lung contusion (LC), hemorrhagic shock and chronic stress (CS), which replicates this persistent-injury associated anemia.^14-16 Following trauma, bone marrow dysfunction has a multifactorial etiology that includes myelo-erythroid reprioritization, decreased growth of bone marrow erythroid progenitor cells, loss of bone marrow progenitor cells from the bone marrow to peripheral blood and sites of injury, an abnormal erythropoietin response, and iron dysregulation.^17,18

Specifically, this rodent model of severe trauma and stress has demonstrated that CS leads to hypercatecholaminemia that drives persistent inflammation and increased hepcidin expression and bone marrow dysfunction.^19,20 Rodents who undergo LCHS but are left to rest for 7 d are physiologically similar to naïve rodents on sacrifice day, whereas rodents who undergo surgery with added daily CS experience this hypercatecholaminemia and bone marrow dysfunction at 7 d from injury.^20,21 The goal of this study is to evaluate long term how an ongoing hypercatecholamine state with persistent inflammation 14 d after initial injury directly contributes to the lack of anemia recovery following severe trauma. We hypothesize that CS and prolonged adrenergic stimulation following injury and hemorrhagic shock contribute to a persistent iron dysregulation that correlated with a lack of anemia recovery.

Materials and methods

Experimental rodent model

Male Sprague-Dawley rats (Charles River, Wilmington, MA) aged 8-9 wk, weighing 275g to 325g were housed in pairs. There was ad lib access to food and water, during daily night and/or day cycles of 12 h each. Female animals were excluded due to estrous cycle variability and its impact after hemorrhagic shock. The animal protocol was approved by the University of Florida Institutional Animal Care and Use Committee.

Rodents were randomly assigned into one of four groups (n = 10/group): (1) naïve controls that underwent daily handling for 14 d; (2) lung contusion followed by hemorrhagic shock followed by daily handling for 13 d (LCHS); (3) LCHS followed by 7 d of CS in a restraint cylinder for 2h/d, followed by 7 d of rest (LCHS/CS-7); (4) LCHS/CS followed by 14 d of chronic restraint stress (LCHS/CS-14) (Fig. 1). For those animals that underwent 14 d of restraint stress, on days 8-14 continuous light was also added to disrupt the day and/or night cycle and prevent habituation. All rodents were weighed daily using a Kent Scientific animal weighing scale (Torrington, CT, USA). Bone marrow, liver and plasma were harvested from all rodents on day 14. Bone marrow, liver and plasma were stored at −80°C.

Fig. 1 –

Model of severe blunt trauma, hemorrhagic shock and chronic stress. (1) naïve controls that underwent daily handling for 14 d; (2) lung contusion followed by hemorrhagic shock followed by daily handling for 13 d (LCHS); (3) LCHS followed by 6 d of chronic stress in a restraint cylinder for 2h/d, followed by 7 d of rest (LCHS/CS-7); (4) LCHS/CS followed by 13 d of chronic restraint stress (LCHS/CS-14).

Lung contusion and hemorrhagic shock

As previously described, after intraperitoneal pentobarbital injection, a unilateral LC was made by a percussive nail gun (Sears Brand, Chicago, IL) applied directly to a 12 mm metal plate that was placed in the right axilla of the rodent.¹⁸ Following LC, the right femoral artery and right internal jugular vein were cannulated using heparinized saline (10 units/mL). Using a continuous BP-2 Digital Blood Pressure Monitor device (Columbus Instruments, Columbus, Ohio) for measurement of mean arterial pressure and heart rate, hemorrhagic shock (HS) was then performed by withdrawing blood to maintain a mean arterial pressure of 30-35 mm Hg for 45 min. Half of the shed blood volume was then reinfused to restore the MAP to normal.

Chronic stress

As previously described, CS consisted of 2 h of restraint in a nose cone rodent cylinder (Kent Scientific Corporation, Torrington, CT, USA) daily for 6 or 13 d after initial injury. Rodents were repositioned in the cylinder every 30 min to prevent habituation and during which alarms sounded for 2 min.¹⁸ The rodents in LCHS/CS groups were restricted from food and water during CS periods.

Bone marrow analysis

As previously described, following removal of the femur, the bone marrow was reamed using a 19g needle and 3 mL syringe containing 1 mL of Iscove’s Modified Dulvecco’s Medium with 2% FBS (Stem Cell, Vancouver, Canada).¹⁶ After centrifugation, the bone marrow supernatant was then decanted. 600 μL of RNA lysis buffer (1 mL of RLT from RNeasy Mini Kit, Qiagen and 10 μL of 2-mercaptoethanol, Sigma) was then added to the cell pellet and homogenized using Pro 200 Homogenizer (PRO Scientific, Oxford CT, USA). The samples were stored at −80°C.

RNA extraction was performed using RNeasy Mini Kit (Qiagen, Hilden, Germany) following the manufacturer’s instructions. RNA was quantified using Biotek Micro-volume plate in Biotek Instruments and a ratio of absorbance at 260/280 nm was taken for RNA purity.

Following synthesis of bone marrow RNA, cDNA synthesis was performed using High-Capacity cDNA Reverse Transcription Kits (Applied Biosystems, Vilnius, Lithuania) as per manufacturer’s instructions. After the cDNA was made, a stock was made by diluting it five times with nuclease free water and this was stored frozen until use. Real-time polymerase chain reaction (PCR) was performed to analyze the expression of bone marrow erythropoietin receptor. This was performed using Brilliant II SYBR Green QPCR Master Mix (Agilent Technologies, Santa Clara, CA) and the Mx3005P qPCR System (Agilent Technologies, Santa Clara, CA), reported as mRNA fold change relative to naïve animals. Primers were designed using Primer3 Web software (Table 1).

Table 1 – –

PCR primers.

	Forward primer	Reverse primer	NCBI reference sequence	Melting point
EPOr	5-GCTCCTATG ACCACCCACAT-3	5-GGTTGCTCAG GACACACTCA-3	NM_017002.2	60°C
Hepcidin	5-GAAGGAAGCG AGACACCAAC-3	5-GAGGTCAGGA CAAGGCTCTT-3	NM_053469.1	60°C
IL-6	5-TGATGGATGC TTCCAAACTG-3	5-GAGCATTGGA AGTTGGGGTA-3	NM_012589.2	60°C
IL-1β	5-CAGGAAGGCA GTGTCACTCA-3	5-AAAGAAGGTG CTTGGGTCCT-3	NM_031512.2	60°C
Beta actin	5-AGCCATGTAC GTAGCCATCC-3	5-CTCTCAGCTG TGGTGGTGAA-3	NM_031144.3	60°C

EPOr = expression of bone marrow erythropoietin receptor.

Liver analysis

Liver RNA was isolated from 30 mg liver tissue using Purelink RNA mini kit (Invitrogen, USA) following the manufacturer’s protocol. Briefly, 650 ul of Lysis buffer with 2-mercaptoethanol (10 ul mercaptoethanol per mL lysis buffer) was added to the liver tissue. Using a rotor-stator homogenizer the tissue was homogenized. The homogenate was passed 5-10 times through a 20-gauge syringe needle. After centrifugation at 16,000 x g for 5 min. The supernatant was then transferred to a clean RNase-free tube. One volume of 70% ethanol was added and vortexed to mix thoroughly. 700 ul of the sample was added to a spin cartridge and centrifuged at 12,000 x g for 15 s. To determine RNA concentrations and purity, the ratio of the UV absorbance at 260 nm and 280 nm was measured. Real-time PCR was then performed to analyze the expression of liver IL-6, IL-1b and hepcidin. This was performed using Brilliant II SYBR Green QPCR Master Mix (Agilent Technologies, Santa Clara, CA) and the Mx3005P qPCR System (Agilent Technologies, Santa Clara, CA), reported as mRNA fold change relative to controls.

Plasma analysis

Plasma IL-6 (Invitrogen, Vienna, Austria), transferrin (Abcam, Cambridge, UK), ferritin (Abcam, Cambridge, UK) and G-CSF (MyBioSource, San Diego, CA) were measured by sandwich enzyme linked immunosorbent assay. All samples were run in duplicate following the manufacturer’s protocol.

Urine analysis

Upon sacrifice, rats in all groups had urine extracted from the bladder using 1 mL needle syringe. Collected urine was transferred into 1.5 mL Eppendorf tubes and frozen at −80°C. Norepinephrine levels were assessed from rat urine samples using the LDN Noradrenaline ELISA Fast Track kit (Nordhorn, Germany) according to manufacturer’s protocol.

Statistical analysis

Statistical analysis was performed using GraphPad Prism version 6.05 (GraphPad Software, La Jolla, CA) to calculate one-way analysis of variance with Bonferroni’s, Sidak’s or Tukey’s multiple comparisons test with a single pooled variance. Data were reported as mean ± standard deviation. Significance was set at α = 0.05.

Results

Prolonged chronic stress impaired anemia recovery

Fourteen days of CS following LCHS led to significantly reduced weight gain when compared to all other groups (Fig. 2A). LCHS/CS-14 led to a 73% and 67% decrease in weight gain when compared to naïve and LCHS rodents respectively (*P < 0.0001 for both). When compared to only 7 d of CS following LCHS, LCHS/CS-14 led to a significant 54% decrease in weight gain (*P= 0.0066). Seven days of CS following LCHS also produced a significant decrease in weight gain when compared to naïve rodents, but this was not significant when compared to LCHS rodents (Fig. 2A).

Fig. 2 –

Prolonged chronic stress following blunt trauma/hemorrhagic shock decreased weight gain and led to hypercatecholaminemia. (A) LCHS/CS-14 stress significantly reduces % weight change when compared to all other groups. (B) LCHS/CS-14 leads to significantly higher levels of urine norepinephrine when compared to LCHS/CS-7 and naïve rodents. *P < 0.05 between groups.

LCHS and LCHS-7 rodents had low urine norepinephrine levels near that of naïve levels on day 14. Fourteen days of CS following LCHS resulted in significantly higher levels of urine norepinephrine when compared to LCHS/CS-7 rodents and naïve rodents, with a 266% and 508% increase respectively (*P = 0.021 and *P = 0.0156 respectively) (Fig. 2B).

LCHS/CS-14 rodents had persistent anemia when compared to both naïve rodents and LCHS rodents with a 7.7 % and 6.1% decrease in hemoglobin respectively (*P = 0.0021 and *P= 0.0287, respectively). LCHS/CS-7 rodents did not have a significant decrease in hemoglobin as compared to LCHS/CS-14, LCHS or naïve animals. There was no significant difference in hemoglobin 2 wk after LCHS when compared to naïve (Hb 13.5 and 14.1 g/dL, respectively).

Prolonged chronic stress led to persistent inflammation

Fourteen days of CS following LCHS/CS-14 led to significantly higher levels of plasma IL-6 when compared to naïve, LCHS and LCHS/CS-7 (9.2 ± 3.8 versus 0.9 ± 0.8, 3.2, 5.1 ± 1.7 pg/mL; *P< 0.0001, *P= 0.0011 and *P= 0.0347 respectively) (Fig. 3A). LCHS/CS-14 led to an 80% increase in plasma IL-6 concentrations when compared to LCHS/CS-7. LCHS/CS-7 had significantly elevated plasma IL-6 levels when compared to naïve rodents (*P= 0.0342) but there was no significant difference between LCHS/CS-7 and LCHS. In addition to systemic IL-6 elevation, there was a significant increase in liver IL-6 expression following LCHS/CS-14 (Fig. 3B). LCHS/CS-14 rodents had a 216% increase in liver IL-6 expression when compared LCHS/CS-7 and LCHS rodents (*P= 0.027).

Fig. 3 –

Prolonged chronic stress led to persistent and worsening systemic inflammation. (A) Plasma IL-6 is significantly increased in LCHS/CS-14 when compared to all other groups. (B) Liver IL-6 expression is significantly increased in LCHS/CS-14 when compared to all other groups. Naïve animals are represented by the dashed line. (C) Liver IL-1b expression is significantly increased in LCHS/CS-14 when compared to naïve and LCHS/CS-7 rodents. Naïve animals are represented by the dashed line. (D) Plasma G-CSF levels were significantly higher in LCHS/CS-14 than LCHS rodents. *P < 0.05 between groups.

Prolonged CS led to significantly higher levels of IL-1 β expression following LCHS/CS-14 when compared to naïve and LCHS/CS-7, with a 50% and 66% increase in expression respectively (*P= 0.0349 and *P= 0.0435, respectively) (Fig. 3C).

In addition, LCHS/CS-14 led to significantly higher levels of plasma G-CSF when compared to LCHS rodents (7.5 ± 4.4 versus 3.0 ± 1.4 pg/mL;*P= 0.0245) (Fig. 3D). Fourteen days following LCHS and LCHS/CS-7, G-CSF levels were similar to naïve.

Prolonged chronic stress led to persistent iron dysregulation

Fourteen days following LCHS, there was a significant decreased in bone marrow erythropoietin receptor expression when compared to naïve and LCHS rodents (*P= 0.0032 and *P= 0.0003 respectively) (Fig. 4A). LCHS/CS-7 rodents also had significantly decreased expression of bone marrow erythropoietin receptor when compared to naïve and LCHS rodents (*P= 0.0107 and *P= 0.0012 respectively), which is consistent with the observation that decreased erythropoietin receptor expression correlated the presence of anemia. There was no significant difference in bone marrow erythropoietin receptor expression following LCHS and naïve on day 14.

Fig. 4 –

Prolonged chronic stress caused persistent iron dysregulation. (A) Bone marrow EPOr expression is significantly suppressed in LCHS/CS-14 when compared to naïve and LCHS rodents. Naïve animals are represented by the dashed line. (B) Liver hepcidin expression is significantly increased in LCHS/CS-14 when compared to all other groups. Naïve animals are represented by the dashed line. (C) Plasma transferrin is significantly decreased in LCHS/CS-14 when compared to naïve and LCHS rodents. (D) Plasma ferritin is significantly decreased in LCHS/CS-14 rodents when compared to naïve and LCHS/CS rodents. *P < 0.05 between groups. EPOr = expression of bone marrow erythropoietin receptor.

Liver hepcidin expression was significantly increased following LCHS/CS-14 when compared to naïve, LCHS and LCHS/CS-7 on day 14 (Fig. 4B). There was an 80% increase in liver hepcidin expression following LCHS/CS-14 group when compared to the LCHS/CS-7 and LCHS (*P= 0.0072 and *P= 0.0441, respectively). There was no significant difference in liver hepcidin expression between LCHS/CS-7 and LCHS or naïve rodents on day 14.

Fourteen days of CS significantly decreased plasma transferrin levels following LCHS/CS-14 when compared to naïve and LCHS, with a 53% and 42% decrease in concentration, respectively (*P= 0.0003 and *P= 0.0174 respectively) (Fig. 4C). LCHS/CS-7 rodents had a significantly decreased plasma transferrin concentration when compared to naïve rodents, but this was not significantly different when compared to LCHS/CS-14.

Fourteen days of CS following LCHS significantly decreased plasma ferritin levels when compared to the naïve and LCHS groups (*P= 0.0122 and *P= 0.0306, respectively) (Fig. 4D). However, there was no significant different in plasma ferritin levels between LCHS/CS-14 and LCHS/CS-7.

Discussion

This study utilized prolonged CS for 2 wk to evaluate the long term the persistence of iron dysfunction and its impact on anemia recovery following injury. These investigations found significant differences in inflammatory and iron dysfunction markers when comparing LCHS/CS-14 rodents to naïve, LCHS and LCHS/CS-7 rodents. Specifically, the differences between LCHS/CS-14 rodents and LCHS/CS-7 rodents demonstrated the role that long term CS played a role in perpetuating reduced weight gain, persistent inflammation, increased hepcidin expression, and iron dysfunction associated with a lack of anemia recovery. Those rodents who had only 7 d of CS following LHCS demonstrated significant recovery on day 14 with reduced inflammation and improving iron dysfunction. No previous studies have demonstrated long term mechanistic insight involving bone marrow dysfunction and anemia recovery.

Douris et al.²² showed that mice who were given a low carbohydrate diet had an increased metabolic rate, which led them to lose weight and suggested that fibroblast growth factor 21 mediated these changes through activation of the sympathetic nervous system. CS alone for 7 d has been shown to reduce weight gain by 50% compared to naïve rodents and following 14 d of CS, there is a 33% reduction in weight gain compared to naïve rodents.²¹ This study demonstrated a significant increase in urine norepinephrine concentrations following LCHS/CS-14. In comparison, LCHS/CS-7 rodents had reduced urine norepinephrine levels demonstrating that 7 d of rest promoted recovery. Naïve rodents gained weight daily and injury alone (LCHS) produced a similar weight gain to that of naïve animals. Similar to a previous study, the addition of CS to LCHS led to decreased weight gain after 7 d when compared to LCHS and naïve rodents.²³ However, prolonged CS for 2 wk led to a persistent decrease in weight gain. However, 7 d of recovery following CS (LCHS/CS-7) allowed rodents to recover and their weight gain was nearly double of the rodents that were stressed for 14 d. This suggests that the persistent hypercatecholaminemia due to the prolonged CS along with persistent inflammation, drives the decrease in weight gain in these rodents following injury.

In a 7 d rodent model of blunt traumatic injury and hemorrhagic shock, the addition of CS for 7 d was associated with urine hypercatecholaminemia and anemia, which was more severe than those who underwent injury and shock alone.¹⁸

Our 14 d model demonstrates anemia in both the LCHS/CS-7 and LCHS/CS-14 groups. The mean hemoglobin on sacrifice day was lower following LCHS/CS-14 than LCHS/CS-7, but this was not statistically or clinically significant. Clinically, it has been shown that hemoglobin levels negatively correlated with injury severity.¹³ Also, decreased hemoglobin levels were found to be closely related to elevated norepinephrine levels in patients with Parkinson’s disease, suggesting that there is a key role for adrenergic stress in patients with anemia.²⁴

In a previous study, Miller et al.²¹ found that CS alone for 7 and 14 d sequentially increased acute phase reactants including norepinephrine, IL-6, tumor necrosis factor-alpha, and C-reactive protein. Fourteen days of restraint stress following LCHS led to significantly and persistently higher levels of systemic IL-6 and G-CSF and increased expression of liver IL-6 and IL-1β. IL-6 is a well-known pro-inflammatory cytokine and is used as a reliable marker for systemic inflammation in patients with moderate to severe trauma.^25,26 IL-6 also has a pathological effect on chronic inflammation and immunity when its synthesis is dysregulated and continual.²⁷ Similarly, Cai et. al²⁸ demonstrated that high-dose G-CSF is associated with prolonged inflammation and can cause severe fibrosis. Elevated G-CSF levels are also associated with prolonged mobilization of hematopoietic progenitor cells into the peripheral blood.²⁹. IL-1β is produced by stimulated monocytes, macrophages and endothelial cells, and is a potent inflammatory cytokine which plays an important role in inflammation in the liver.^30,31 Together, these plasma and liver expression findings suggests that prolonged CS led to persistent systemic inflammation. Allowing 7 d of rest following LCHS/CS-7, there was reduced inflammation similar to injury alone and naïve animals. IL-1β and IL-6 have been shown to contribute to anemia through erythropoietin resistance, myelosuppression, and decreased erythrocyte survival.³² In addition, a recent randomized trial demonstrating that inflammation inhibition through IL-1β/IL-6 signaling pathway reduced the incidence of anemia and improved hemoglobin levels.³³

Hepcidin has been shown to contribute to dysregulated iron trafficking after severe trauma, and its elevation led to a relative iron deficiency that is also seen in the anemia of inflammation.^19,20,34 Hepcidin expression is strongly stimulated during times of inflammation or infection, largely by IL-6 which was also shown in this study.¹⁹ Exogenous hepcidin injection in a murine model was associated with decreased iron bioavailability and iron-restricted erythropoiesis.³⁵ When hepcidin is deleted, there is an attenuation of anemia development during acute inflammation.¹⁹ Loftus et al.¹⁷ described the iron dysregulation pathway that occurs in severely injured trauma patients and highlighted the role that hepcidin played in decreasing iron bioavailability and blunting the response of bone marrow erythropoietin receptor. Elevated IL-6 stimulates an increase in systemic hepcidin, which then causes ferroportin to be internalized, thereby decreasing mobilization of iron stores.^19,33,34 This study is consistent with those findings, 14 d of CS following injury led to prolonged IL-6 elevation, increased liver expression of hepcidin and a reduction in the expression of the erythropoietin receptor in rodents that had persistent anemia.

In addition to the systemic increase in IL-6, this study found increased expression of liver IL-6 and hepcidin following LCHS/CS-14. Yet, in those animals that had 7 d of rest, following LCHS/CS-7 plasma IL-6 and liver expression of IL-6 and hepcidin were reduced similar to that of injury alone and naïve animals. Prolonged CS contributed to persistent inflammation and continued iron dysfunction after trauma. With increased expression of hepcidin following LCHS/CS-14 there was decreased iron bioavailability reflected in the decreased plasma ferritin and transferrin levels. Plasma transferrin has been shown to play a key role in transporting iron to erythroid progenitor cells for uptake.³⁶ These findings together suggest that iron homeostasis is disrupted on multiple levels by prolonged CS, increased IL-6 and hepcidin, with blunted erythropoietin receptor expression, and reduced plasma ferritin and transferrin levels, all of which contribute to decreased iron availability for uptake into erythroid progenitor cells and impaired erythropoiesis with no evidence of anemia recovery after 14 d.

The clinical translation of these rodent findings depends on effectively recapitulating neuroendocrine activation and inflammation experienced by severely injured trauma patients who are admitted to the intensive care unit. Rodent models have poor track record for translation in human studies due to the complex pathophysiology of human disease processes. Many trauma animal models utilize early timepoints that may be well-suited for mortality studies, but this does not allow for the mechanistic understanding of the pathophysiology. This unique rodent model has been developed to recapitulate severe traumatic injury followed by an intensive care unit stay. In addition to injury and hemorrhagic shock, the addition of 14 d of CS to simulate an intensive care unit environment with alarms, restraint and repositioning has allowed for long term study of injury and recovery processes. In this study, female rodents were excluded due to the known influence of sex hormones on outcomes after hemorrhagic shock. Future studies will include female rodents to evaluate the impact of estrous state on hemorrhagic shock and injury recovery to mitigate this limitation.

Conclusions

Prolonged CS for 2 wk in a model of severe traumatic injury not only leads to persistent hypercatecholaminemia, but also worsened systemic inflammation and impaired iron function, thereby contributing to a lack of anemia recovery 2 wk after injury. This is confirmed by the observation that LCHS/CS-7 rodents, who then recovered for 7 d are more physiologically similar to injury alone and naïve rodents. Further studies are warranted to determine how we can reduce CS in our trauma population, in order to reduce the burden of inflammation as well as decrease hepcidin to improve their anemia recovery and subsequent functional outcomes.