Anemia in Chronic Obstructive Pulmonary Disease and the Potential Role of Iron Deficiency

Abstract

The purpose of this review is to evaluate the role of anemia on patient outcomes in chronic obstructive pulmonary disease (COPD), the potential contribution that low iron stores may play in this process, and possible treatment considerations. A review of research studies found that anemia is associated with declining functional outcomes, increased health care utilization and costs, and increased mortality in COPD. Associations exist between reduced iron intake and progression of COPD and in reduction of iron status with declining lung function. Currently data are limited on the effects of either treating anemia or utilizing iron supplementation in anemic COPD patients. If iron supplementation might therefore reverse some of the declines that patients experience, then routine screening and treatment may turn out to be an effective, simple and inexpensive intervention. Iron supplementation models utilized in other inflammatory-related disease states were reviewed as a possible starting point to evaluate treatment options in COPD. Future research can be directed to establish best practice standards for the use of iron supplementation in COPD.

Keywords:

• health care cost
• inflammation
• mortality
• quality of life

Introduction

Overall COPD affects approximately 6.3% of adults in the United States and is known as a significant contributor to nationwide mortality rates (1). Anemia has recently been linked to increased disease burden and mortality in COPD and is estimated to affect 10–30% of COPD patients (2–16). Iron deficiency may play a significant role in the progression of anemia within this population and supplementation may show promise in anemic COPD patients. Currently, research is inconsistent on how low iron stores are defined and is limited on the effects of iron supplementation in COPD.

The majority of patients with COPD also suffer from co-morbidities. A key component in both COPD and the primary co-morbidities observed is an underlying chronic inflammatory process (17, 18). This inflammatory process is a key consideration in evaluating anemic COPD patients. The two primary types of anemia seen in COPD are anemia of chronic disease (ACD) and iron deficiency anemia (IDA) with most cases attributed to ACD (2). Both of these are related to alterations in iron balance. Though the two anemias can coexist, there is little data available on the prevalence of mixed ACD with iron deficiency (2, 8, 14).

Traditionally, IDA in COPD has been defined by the presence of anemia with a low ferritin, yet due to the effects of inflammation on iron metabolism many believe adequate ferritin should be defined differently in the presence of chronic inflammation (2, 14, 19–21). Definitions have been proposed to identify this subgroup in one study assessing anemia in COPD and in other disease states (14, 19–21). Silverberg and colleagues (14) applied this criterion to patients hospitalized with acute exacerbation of COPD. Of the anemic patients studied, only 38% had iron studies conducted. Of those, every patient met criteria for either IDA or ACD with functional iron deficiency. Researchers also noted that none of the patients had received iron either during their hospital stay or at discharge.

Although current data are limited, iron supplementation may prove to be relatively inexpensive and an effective treatment option for anemic COPD patients. Research conducted in other inflammatory-related disease states, and one study of COPD with impaired renal function, have shown improvement in patient outcomes (14, 19–21). The objectives of this review are to identify the effects that anemia has on patient outcomes in COPD, to explore if low iron stores may be a contributing factor, and to evaluate the potential benefits of treatment. The roles of inflammation and erythropoietin, the potential effects of both iron and anemia on oxidative stress and inflammation, and methods of iron supplementation, will be reviewed. A PubMed search was conducted to identify relevant literature. Search terms included “COPD,” “anemia,” “anaemia,” “iron,” “iron deficiency,” “mortality” and “quality of life.”

The Connection between Inflammation, Iron Deficiency and Anemia

Recent attention to the effects of inflammation on iron metabolism has identified that ACD can result in a functional iron deficiency (22–32). Inflammatory proteins have been shown to stimulate hepatic production of hepcidin, a protein involved in decreasing iron absorption and increasing iron storage (22). Elevation in hepcidin resulting from chronic inflammation may not respond appropriately to declining iron stores and may over time induce a functional iron deficiency.

In healthy individuals, hepcidin is a crucial protein involved in preventing iron toxicity. Hepcidin production increases when iron intake is high and subsequently binds to and degrades the iron export protein, ferroportin. With declining levels of ferroportin, the body's ability to mobilize iron is limited. Iron becomes trapped in the macrophages, hepatocytes and duodenal enterocytes causing serum iron to fall and ferritin levels to rise. As the iron available for physiologic use declines, hepcidin levels decline resulting in increased ferroportin. Ferroportin then mobilizes iron from ferritin and duodenal enterocytes to support adequate serum levels (31, 32).

In addition to excessive iron, inflammatory proteins, specifically interleukin-6, appear to be a key trigger for hepcidin production (22, 30–32). The rise of the hormone hepcidin resulting from inflammation causes the same end effects of iron becoming trapped as ferritin and reduced ability to mobilize dietary iron from the duodenal enterocytes. As inflammation becomes chronic, hepcidin levels may remain inappropriately elevated relative the amount of usable iron in the body. This prolonged maladaptive response can result in ACD with functional iron deficiency. In these patients intravenous iron administration is preferred over oral supplementation due to the decreased ability to transport the enteral iron into the plasma (31).

Measures of Iron Status and Functional Iron Deficiency .

Traditional measures of iron status have not been as reliable in identifying functional iron deficiencies. Transferrin saturation, total iron binding capacity (TIBC) and ferritin are all commonly measured when deficiency is suspected; however each of these measures are known to be altered with inflammation. Without established measures for functional iron deficiency, most studies in populations with chronic inflammation have focused on redefining acceptable levels of these markers.

In simple iron deficiency there is an increase in transferrin production to help mobilize and deliver iron, a resulting decrease in free iron and ferritin, increase in available transferrin binding sites and reduction in transferrin saturation. Inflammation causes an overall increase in ferritin, decrease in free iron and decrease in transferrin. This overall decrease in transferrin decreases potential binding sites in the blood below those for healthy individuals (30). Since transferrin is the primary transport protein of iron in the blood, TIBC is an indirect reflection of transferrin levels (33). This decreases the applicability of TIBC in inflammation.

Measures of soluble transferrin receptor (sTFR) appear to hold promise in evaluating iron status during inflammation (33). sTFR is a truncated form of the membrane transferrin receptor that is released from cells needing iron. Cellular need and therefore sTFR is less affected by inflammation, making sTFR values a unique indicator for iron status. Putting this measure into practice has proven more difficult and currently there are not international standards in place for laboratory measurement or evaluation (34). To date there is not enough evidence to benchmark at which levels of sTFR iron administration should be considered in patients with chronic inflammation (29–32). Additionally, criteria evaluated in other disease states using traditional measures of iron stores have shown improvement in patient outcomes and their correlating sTFR levels in these populations are not known. It is possible that the addition of sTFR may alter treatment groups therefore altering patient outcomes.

Other criteria have also been suggested for functional iron deficiency (34). Percent hypochromic red cells (%HRC) have been found to be effective in the evaluation of patients with chronic renal failure, those with advanced acquired immunodeficiency syndrome and patients with rheumatoid arthritis. Additionally the reticulocyte hemoglobin content (CHr) has been proposed and found to be strong predictor for iron depletion. Since some patients above the traditional threshold for CHr respond well to iron supplementation, a higher cut-off value has also been proposed. Measures used to evaluate iron status in other inflammatory disease states and one study evaluating COPD are listed in Table 1. Standards for %HRC and CHr are also included for comparison.

Table 1. Possible indicators of iron deficiency anemia, anemia of chronic disease, and combined anemia of chronic disease with iron deficiency (14,19–21,34)

Indicator	IDA	ACD	ACD with Functional Iron Deficiency
Hemoglobin	< 13 males	< 13 males	< 13 males
	< 12 females	< 12 females	< 12 females
%Transferrin Saturation	< 20%	> 20%	< 20%
Ferritin	< 100 ng/mL	>300 ng/mL	100–300 ng/mL
%HRC	≥ 6%	< 6%	≥ 6%
CHr	< 29 pg	>29 pg	< 29 pg

Abbreviations: IDA (iron deficiency anemia), ACD (anemia of chronic disease), %HRC (percent hypochromic red cells), CHr (reticulocyte hemoglobin content).

Associations of Anemia and Iron Indicators with Disease Outcomes .

Many studies have demonstrated increased mortality, decreased quality of life and functional capacity in anemic patients with COPD when compared to COPD patients without anemia. In addition to direct measures of self-reported quality of life, anemia has been correlated to increased hospital admissions and both health care costs and utilization. Overall, patients with anemia tend to be sicker, need medical attention more often, require more treatments, and die sooner than their non-anemic counterparts (Table 2). Although there are many potential benefits to long-term disease burden and mortality rates that may be achieved with correcting anemia in COPD patients, no studies have been conducted evaluating long-term outcomes.

Table 2. The effects of anemia on COPD outcomes (2–12)

Author	Sample Characteristics	Anemia Prevalence	Outcomes Associated with Anemia	Other
Krishnan et al. (4)	General population, 35–79 years (n = 2,704; 495 had moderate/severe COPD)	7.5% of COPD participants (Hb <12 g/dL F, < 13 g/dL M)	Reduction in quality-of-life score for physical component summary (p=0.003).	Sampling from general population and low proportion with anemia may have indicated a healthier study population.
Schnecke-npointner et al. (2)	Patients with chronic respiratory failure being followed for intermittent home mechanical ventilation (n = 185, 85 with COPD)	17.6% COPD participants (n = 15; Hb <12 g/dL F, <13 g/dL M)	Hb and TSAT ≥20% independently associated with improved health-related quality-of-life scores (Hb p=.036, TSAT p=.005) and dyspnea (Hb p=.005, TSAT p=.006). Multivariate analysis: serum iron and TSAT were independent prognostic indicators (p < .05 each)	Researchers evaluated IDA (n = 9), ACD (n = 13) and mixed ACD+IDA (n = 1) within study population. IDA defined as TSAT<20% and ferritin< 35 ng/mL; ACD+IDA as TSAT < 20%, ferritin 35–199 mg/mL with sTFR/log ferritin ≥ 1.5; ACD as TSAT < 20%, ferritin ≥ 200 ng/mL or TSAT < 20%, Ferritin 35—199 ng/mL and sTFR/log ferritin < 1.5.
Cote et al. (6)	Clinically stable COPD outpatients (n = 683, of which 677 included in analysis)	17% Hb < 13 g/dL	Multivariate regression analysis: anemia independent predictor of dyspnea (p < 0.05) and lower 6-minute walk distance (p=0.001)	Anemia defined similar in M/F due to debate on appropriate postmenopausal F Hb levels and low proportion of F (4%). Hb differed among survivors (p < 0.01) but with multiple regression anemia was not an independent mortality predictor.
Chambellan et al. (7)	Entered into the French homecare database for chronic respiratory insufficiency and being followed for home oxygen support, mechanical ventilation or continuous positive airway pressure (n = 2,524)	12.6% M (Hct < 39%) and 8.2% F (Hct <36%)	Hct associated with FEV1 percentage of predicted (r = -0.054, p=0.006) and FEV1/VC (r = -0.084, p < 0.001) Multivariate regression: Hct independent predictor of survival (p < 0.001) hospital admission rates (p < 0.002) and cumulative duration of hospitalization (p < 0.001). Average 3-year survival 24% when hct < 35% (95% CI 16–33%) and 70% when hct ≥ 55% (95% CI 63–76%).	Researchers had limited access to co-morbidity data. Only a small percentage with Hct > 55% (8.4%).
Boutou et al. (8)	Stable COPD outpatients (n = 283 for the prevalence study, 27 cases with ACD matched to 27 controls for exercise testing)	10% ACD (Hb < 12 g/dL F, <13 g/dL M, ferritin >30 mg/mL, TIBC< 250 ug/dL, TSAT 15-50%)	ACD was associated with a higher dyspnea score (p < 0.001), lower VO2% predicted (p=0.005), peak work rate (p=0.014, percent predicted p=0.023) and peak VO2/heart rate (percent predicted p=0.01, ml/beat p=0.023)	Patient stability defined as: no exacerbation, hospital admission, respiratory infection, or change in medication for 3 months prior to entering study
Boutou et al. (9)	Stable COPD outpatients (n = 294)	15.6% (Hb <13 g/dL)	Anemia independent predictor of mortality (HR 1.87, 95% CI 1.06–3.29) Anemia associated with shorter median survival (p=0.035)	Anemia defined similar in M/F due to debate on appropriate postmenopausal F Hb levels Mortality associated to anemia, not Hb level
Halpern et al. (3)	U.S. Medicare database claims from 1997–2001 (n = 132,424)	21% (ICD-9 code or receipt of transfusion)	Medicare payments greater for anemic patients (∃1,466 vs. ∃649, p < 0.001) Multivariate analysis: Anemia associated with ∃3,582 increase in per patient annual reimbursements. Anemia associated with higher mortality rate (p < 0.001), ICU admission (p < 0.001), ventilation episodes (p < 0.001), and nursing home admission (p < 0.001).	Patients with other possible causes of anemia were excluded by ICD-9 coding (end stage renal disease, cancer, gastrointestinal bleeding, known nutritional or hereditary anemias) Possibility of coding bias
Shorr et al. (11)	Retrospective cohort of claims made by an integrated healthcare system in the Mid-Western US from 1997-2005 (n=2,404)	33% (Hb < 12 g/dL F, < 13 g/dL M twice in 14–356 days)	Higher annual costs for anemic COPD patients (∃17,240 vs. ∃6,492, p < 0.001), greater healthcare encounters (p=0.0001), hospitalization rates (p=0.0001) and days of duration (p=0.0001). Multiple regression analysis: anemia increased annual cost by ∃7,929 per patient (p < 0.0001)	Anemia was verified in the records; however, identification of COPD was dependent on ICD-9 coding
Ozyilmaz et al. (12)	Patients admitted for COPD exacerbation to out- or inpatient clinics (n = 107)	n/a	Multivariate logistic regression analysis: Hct < 41% independent predictor of frequent exacerbations (p=0.009), Hct negatively correlated with frequent severe exacerbations (p=0.014) Multivariate analysis: Hct < 41% independent predictor of hospital readmission within 2 months (p=0.002)	A cutoff of Hct < 41% was used because within the study population it demonstrated the greatest sensitivity/specificity.
Kollert et al. (5)	Patients being treated with home noninvasive ventilation (n = 309)	14.9% (Hb < 12 g/dL F, <13 g/dL M)	Multivariate analysis: Improvement in survival associated with Hb of 14.3 mg/dL in F, 15.1 mg/dL in M (p < 0.001), these levels independently predicted survival (p=0.03) Hb associated with long term survival (p < .001) Higher survival rate in polycythemic patients vs. anemic patients (p=0.01)	18.1% had polycythemia (Hb ≥ 15 g/dL F, ≥17 g/dL M)
Martinez-Rivera et al. (10)	Patients hospitalized with acute COPD exacerbation (n = 117)	33% (Hb < 13 g/dL)	Anemia independently associated with increased mortality (p=0.000, χ^2 test), shorter mean survival (p < 0.000, log rank test) Anemic patients had a RR of death of 5.9 IC 95% 1.9–18.9, p=0.003).	Exclusion criteria included patients requiring immediate ICU monitoring and/or invasive or noninvasive ventilation. Anemia defined similar in M/F due to debate on appropriate postmenopausal F Hb levels and low proportion of F (7%).

Abbreviations: COPD: chronic obstructive pulmonary disease, Hb: hemoglobin, TSAT: transferrin saturation, Hct: hematocrit, sTFR: serum transferrin receptor, TIBC: total iron binding capacity, M: males, F: females, IDA: iron deficiency anemia, ACD: anemia of chronic disease, FEV₁: forced expiratory volume in 1 second, VC: vital capacity, VO₂: oxygen uptake, U.S.: United States, ICU: intensive care unit.

Two studies have shown a correlation with anemia and quality-of-life measures in COPD. One study conducted a post hoc analysis of 2,704 randomly sampled adults evaluating how anemia affected health-related quality-of-life scores (4). Of the study population, 495 participants had COPD and 7.5% of the COPD participants had anemia. Among the anemic participants, there was a significant reduction of quality-of-life scores in the area of the physical component summary. Other areas did not reach significance when adjusted for co-morbidities. Unfortunately, the power of the study was limited by the relatively low prevalence of anemia. Schneckenpointner and colleagues (2) studied 185 patients with chronic respiratory failure and found that both hemoglobin and transferrin saturation were independently associated with improved health-related quality-of-life scores in COPD patients.

Some of the variation in results between the two studies may in part be due to the populations sampled. The first study sampled people with COPD from the general population while the second study enrolled participants admitted for follow-up of intermittent home mechanical ventilation therapy. It is likely that the second group may have represented a sicker population to sample from, which would also relate to the higher percentage of anemia seen in the second study population. Further studies are needed in this area to clarify how much of an impact anemia may have on quality-of-life measures in COPD.

Another area that has been evaluated is the effect of anemia on direct measures of lung function. A study on 683 stable COPD outpatients demonstrated that the presence of anemia independently predicted dyspnea and lower 6-minute walk distance (6). The ANTADIR study was conducted on 2,524 patients with chronic respiratory failure being followed for intermittent home mechanical ventilation (7). In regards to dyspnea, researchers found a weak negative correlation between hematocrit and forced expiratory volume in 1 second (FEV₁) percent predicted and hematocrit and FEV₁/vital capacity. Another study evaluating functional capacity in anemic patients with COPD found ACD negatively impacted on dyspnea and circulatory efficiency during exercise (8). In the study by Schneckenpointner and colleagues (2), researchers found hemoglobin and transferrin saturation to be independently associated with dyspnea. Current research on the effects of anemia on functional capacity in COPD consistently demonstrates decreased measures of lung function associated directly with indicators of anemia.

Iron status has also been associated with measures of lung function (Table 3). One population-based cross-sectional study evaluated serum iron and lung function and found that although there was not a clear dose-response relationship, serum iron was positively associated with FEV₁ (35). Additionally Schneckenpointner and colleagues (2) found that among clinically stable COPD patients, lower percent transferrin saturation was associated with worsening dyspnea. Another indirect approach taken to look at the relationship between iron and lung function has been evaluating iron intake. One study compared iron intake with percent predicted FEV₁ in symptomatic elderly smokers and stable COPD patients and demonstrated a positive correlation between the two measures (36). A study evaluating iron intake and risk for COPD showed reduced iron intake was associated with increased risk of COPD and with percent predicted FEV₁, though the magnitude of the correlation was low (37).

Table 3. Studies of iron in COPD or on lung function (35–38)

Author	Study Design	Findings	Comments
McKeever et al. (35)	Data from the Third NHANES (US population cross-sectional study). Measured serum markers of 21 nutrients and FEV1	Increased serum iron correlated with increased FEV1 (p=0.0054).	Serum iron may also influenced by inflammation.
Obase et al. (36)	13 stable COPD patients, 10 smokers with COPD symptoms, 27 healthy controls, 70 years/older. Respiratory function and diet intake was evaluated.	% predicted FEV1 correlated with iron intake in the COPD group (p=0.03) Symptomatic smokers and those with COPD had lower iron intake (p < 0.0001)	Diet intake evaluated over 24 hours.
Hirayama et al. (37)	278 COPD patients ages 50–75 years, control group 340 community dwelling adults. Respiratory function was evaluated with intake of 6 minerals.	Lower COPD risk with higher energy adjusted intakes of iron (OR 0.68, 95% CI 0.39–0.99, test for linear trend p=0.01) Iron intake associated with % predicted FEV1 (p=0.001)	Food frequency questionnaire used for dietary evaluation.
Smith et al. (38)	16 healthy volunteers with normal iron status were exposed to hypoxia either with or without prior iron supplementation.	Iron supplementation prior to hypoxic insult resulted in a lower rise in baseline PASP with prolonged hypoxia (p < 0.001) and attenuated the increase in PASP response to acute hypoxia (p=0.002)	Increases in PASP resulting from disease related hypoxia can be seen in COPD.

Abbreviations: COPD: chronic obstructive pulmonary disease, NHANES: National Health and Nutrition Examination Survey, U.S.: United States, FEV₁: forced expiratory volume in 1 second, PASP: pulmonary arterial systolic pressure.

Data describing the whether the patients were in stable or exacerbated states were not reported. One study evaluating healthy subjects exposed to hypoxia found that supplementing iron prior to hypoxic insult reduced the rise observed in pulmonary arterial systolic pressure (38). The researchers speculated that diseases demonstrating a similar hypoxic insult may benefit from ensuring adequate iron status. Current research indicates a relationship exists between iron and lung function, but has yet to clearly define the role that iron may play in this process.

Another consideration when evaluating quality of life may be health care utilization. In a study evaluating U.S. Medicare database claims, researchers found that anemia was associated with an additional ∃3,582 increase per patient in reimbursements, higher rates of intensive care unit admissions, ventilator episodes and nursing home admissions (3). One study evaluated per patient costs for COPD in a large health care system (11). Researchers found annual costs for COPD patients with anemia were significantly more and healthcare utilization significantly greater than in those without anemia. In their multiple regression analysis, anemia was associated with and increased cost of ∃7,929 per patient. Ozyilmaz and colleagues (12) looked into potentially modifiable risk factors for acute exacerbation of COPD. Acute exacerbations requiring hospitalization are considered a prognostic factor and leading cause of fatal events. They evaluated 107 patients with COPD and found that a hematocrit of < 41% was an independent risk factor for frequent severe exacerbations and readmission. Additionally, in the ANTADIR study researchers found hematocrit to be an independent predictor of hospital admission rates and cumulative duration of hospitalization (7). Overall, anemic COPD patients have higher acute care hospital readmission rates and overall treatment costs.

Along with declines in quality of life and measures of lung function, anemic patients with COPD demonstrate higher mortality rates. The ANTADIR study demonstrated hematocrit to be an independent predictor of survival in their study population (7). They found that the three-year survival rate was 24% when the hematocrit was < 35%, and 70% when the hematocrit was ≥ 55%. Another study evaluated hemoglobin levels in 309 patients with COPD and found significant improvement in survival in patients with hemoglobin levels of 14.3 for females and 15.1 for males (5). The study evaluating U.S. Medicare claims also showed that having COPD with anemia was associated with significantly higher mortality rates, and the Schneckenpointner and colleagues’ study demonstrated that serum iron levels and transferrin saturation were independent prognostic indicators (2, 3).

A study on 294 stable COPD outpatients showed anemia was an independent predictor of mortality, associated with shorter median survival and 3- and 5-year mortality rates (9). One study evaluating 117 patients hospitalized for acute exacerbation of COPD found hemoglobin values of <13 mg/dL were independently associated with increased mortality risk and an RR of death of 5.9 for anemic patients (10). In this study, 43.6% of anemic patients died in the one-year follow-up period while only 11.8% of patients with hemoglobin >13 mg/dl died. Whether a correction of anemia alone could improve these patient outcomes has yet to be evaluated. Because of limited information in this area, it is also possible that the disease process causing the anemia and not the anemia itself is the major contributing factor to increased mortality rates.

Anemia and Erythropoietin in the Treatment of COPD

Some treatment strategies for anemia in other inflammatory-related disease states have used a combination of iron and erythropoietin stimulating agents (ESAs) (14, 20, 21). Studies evaluating anemia in COPD have revealed an altered production of or response to erythropoietin among some patients. The relationship between iron, hypoxia, inflammation and erythropoietin is complex with studies showing mixed results. In the setting of hypoxia, renal production of erythropoietin is stimulated, resulting in increased red blood cell production from the bone marrow. These responses appear blunted among some patients with COPD at either the level of the kidney or bone marrow. This blunted response may be related to inflammation or other causal factors yet to be identified (23, 24). The role of ESAs in treating anemia in COPD is currently unclear.

Some studies have demonstrated an overall resistance to elevated erythropoietin levels in anemic patients. One study conducted on 101 COPD patients found that the anemic patients had increased erythropoietin levels and an increased inflammatory response (23). Another study was conducted on 54 stable COPD patients who were evaluated for markers of inflammation (24). Those with anemia of chronic disease had significantly higher levels of inflammatory protein markers than controls and also demonstrated significantly higher erythropoietin levels. One study found that during acute exacerbation of COPD erythropoietin levels were increased and hemoglobin levels were decreased (39). Additionally, they identified a correlation of this response with increased systemic inflammation, specifically with interleukin-6. With resolution of the patients acute exacerbation researchers observed that the relationship between hemoglobin and erythropoietin became positive. These studies concluded that the anemic patients had resistance to erythropoietin, and this may play a role in preventing the correction of the anemia. The response seen could be indicative of an altered response of the bone marrow to erythropoietin that may be due to chronic inflammation or possibly the resulting poor supply of iron blunting the response to erythropoietin (23, 24).

Other studies assessing erythropoietin in COPD have demonstrated an altered response in erythropoietin to hypoxia among some patients. A study of 29 COPD patients with hypoxia compared to 18 patients with hypoxia from diffuse idiopathic pulmonary fibrosis found an inconsistent erythropoietin response to hypoxia among the COPD group (40). Another study evaluated 57 consecutive patients with COPD and chronic hypoxia admitted with acute exacerbation for their response to correction of the hypoxia (41). Although mean erythropoietin levels were above the normal range, there was a subset of patients (36.8%) who initially presented with low erythropoietin levels, which then significantly increased on day two after correction of the hypoxia. Sala and colleagues (42) evaluated erythropoietin levels in stable COPD patients and those with acute exacerbation of COPD. They found that in the exacerbation group erythropoietin levels were significantly lower and inflammatory markers increased when compared with the stable group. The researchers concluded that some patients with COPD have an altered production of erythropoietin in response to hypoxia that correlated with an increase in systemic inflammatory markers. These results may indicate a resistance at the level of the kidney.

The complex picture of these studies indicates there might be a place for ESAs in the correction of anemia among patients with COPD. A combined approach with ESAs and intravenous (IV) iron supplementation has been used in patients with congestive heart failure and has demonstrated significant benefit to patient outcomes (20, 21). Further research is needed in the area of iron supplementation or iron supplementation with ESAs to evaluate which may be the most effective in correcting anemia and improving patient outcomes.

Treatment of Anemia in Patients with COPD

Despite the knowledge that anemia is a potentially modifiable risk factor for poorer patient outcomes in COPD very few studies have evaluated the effects of treatment (Table 4). There have been two studies that evaluated blood transfusions to correct anemia. One was a series of case studies on patients with COPD and difficulty weaning from ventilator support (43). They found that after correction of anemia by blood transfusion they were able to wean the patients from mechanical ventilation. Another study was done on anemic patients with COPD and found that minute ventilation and work of breathing were associated with an improvement in hemoglobin seen after receiving blood transfusions (44). Provision of anabolic steroids has also been evaluated in one study of patients with COPD (45). Results showed improvements in inspiratory mouth pressure and peak workload to be associated with improvements in hemoglobin. These studies show that positive short-term results may be seen with improved levels of hemoglobin among anemic patients with COPD.

Table 4. Studies on correction of anemia in COPD (14,43–45)

Author	Study Design	Treatment	Findings
Silverberg et al. (14)	Prevalence study: 107 patients hospitalized for acute exacerbation of COPD Treatment study: 11 of 12 patients with COPD/chronic renal failure meeting criteria for iron deficiency	None IV Iron and ESA (Recormon)	43.9% were anemic. Of the anemic patients 18 had iron studies completed during admission, all 18 met criteria for iron deficiency. None received treatment during hospitalization or at discharge. Change in Hb correlated to the change in severity of dyspnea (VAS) (r = 0.71, p=0.009). Change in Hct correlated to the change in VAS (r = 0.8, p=0.0014).
Schönhofer et al. (43)	Case studies: 5 anemic patients with COPD and difficulty weaning from ventilator support	PRBC provided to achieve Hb ≥ 12 g.dL^-1	After receiving PRBC to correct the anemia, patients were able to wean from ventilator support.
Schönhofer et al. (44)	20 anemic adults (Hb < 11 g/dL) 10 with severe COPD, 10 without lung disease	1 unit PRBC for each g/dL Hb was below 11–12 g/dL	Reduction in mean minute ventilation (p < 0.0001), decreased work of breathing (p < 0.0001), increased capillary PCO2 (p < 0.05) decreased PO2 (p < 0.05). No change seen in lung function measures for the control group.
Creutzberg et al. (45)	Double blind, randomly assigned 63 COPD patients to either treatment or control group	On days 1, 15, 29, 43 injection of nandrolone decanoate (50 mg) in arachis oil or arachis oil alone	Treatment group had a tendency towards increases in Hct (p=0.1), Hb (p=0.15) and erythropoietin (p=0.15). Change in maximal inspiratory mouth pressure (r = 0.29, p=0.039) and peak workload (r = 0.33, p=0.019) significantly associated with change in Hb independent of change in FFM. Change in work of lower extremities associated with change in erythropoietin independent of change in FFM (r = .37, p=0.015).

Abbreviations: COPD: chronic obstructive pulmonary disease, Hb: hemoglobin, Hct: hematocrit, VAS: visual analogue scale (measures severity of dyspnea), PRBC: packed red blood cells, FFM: fat free mass.

To date, only one study conducted by Silverberg and colleagues has been published on the effects of iron and ESAs on anemia and patient outcomes in COPD (14). They evaluated patients attending a clinic for treating anemia in chronic kidney disease and congestive heart failure. They retrospectively identified 12 patients attending their clinic who also had COPD. Of these participants, 91.7% met the study criteria for either IDA or ACD with iron deficiency. Significant improvements were seen in hemoglobin, hematocrit, red blood cell count, serum ferritin, and % transferrin saturation after treatment. Participants also showed a significant improvement in their self-assessed shortness of breath Visual Analogue Scale (VAS).

Although the baseline VAS did not correlate with baseline hemoglobin or hematocrit, there was a significant correlation with the change in either hemoglobin or hematocrit with the improvement in VAS. The generalizability of their results is limited by the small sample size and the prevalence of chronic kidney disease within the population examined. Since iron supplementation and ESAs have demonstrated significant benefits in chronic kidney disease, it is difficult to extrapolate if the same results would be seen in a population without chronic kidney disease. The study demonstrates how improvements in hemoglobin and hematocrit result in improved self-reported lung function and indicates the need for further research among patients with COPD.

The Effects of Iron and Anemia on Oxidative Stress

Iron has an established role in increasing oxidative stress and inflammation. In the body, free and poorly liganded iron can react to produce a hydroxyl radical (46–48). The oxidative damage from this radical has been shown to damage membranes, proteins, and DNA and contribute to increasing oxidative stress, inflammation, immune dysfunction, and tissue injury (46, 47, 49). To prevent this damage, most circulating iron is bound to transferrin, and cellular iron to ferritin (46, 47). Administration of IV iron preparations has shown to increase levels of poorly-liganded iron, markers of oxidative stress, endothelial dysfunction, inflammation, and impaired immunity (50, 51). Over time, excessive IV iron supplementation and iron overload may contribute to increased risk of cardiovascular disease and infection (46, 50).

In ACD with functional iron deficiency, ferritin levels tend to increase while free and transferrin-bound iron levels are reduced (14, 19–21). One study found the ferritin content of alveolar macrophages in patients with COPD to be elevated (52). With further examination, they revealed an increased expression of transferrin and ferritin (used to transport iron into the cells and storage) without changes in the expression of ferroportin (used to transport iron out of the cells). These results are consistent with the shifts in iron metabolism seen with ACD. While ferritin helps prevent the oxidative damage associated with free iron, some research has also implicated ferritin in contributing to lipid peroxidation (49). Although systemic inflammation contributes to elevating ferritin levels, no conclusive guidelines are available on what levels of ferritin are considered optimal in patients with chronic inflammation. Additionally, given that the degree of inflammation can significantly affect ferritin response, optimal levels may vary by the patient's condition and degree or length of time that inflammation has been present.

Although iron supplementation can increase oxidative stress, consideration also should be given to the systemic effects of iron deficiency. In addition to iron's role in RBC production and oxygen transport, iron is a cofactor for many enzymes (53). A deficiency of iron for these reactions can affect immunologic and inflammatory defenses. IDA has also been tied to free radical production, lipid peroxidation and altered erythrocyte antioxidant enzyme activity (54, 55). In a study evaluating lipid peroxidation during repletion of body iron stores in participants with IDA, researchers found that lipid peroxidation decreased with the normalization of hemoglobin and saturation of body iron stores (54).

Another study examined how iron supplementation affects erythrocyte antioxidant defenses in participants with IDA (55). In the study researchers found that although IV iron improved levels of erythrocyte superoxide dismutase, levels of glutathione peroxidase were decreased. The addition of oral vitamin E to the IV iron therapy regimen mitigated some of this effect. It is possible that some of the potentially negative effects of IV iron supplementation may in part be balanced out by correction of the underlying iron deficiency.

Chronic inflammatory processes are also associated with the anemia seen in patients undergoing hemodialysis and in congestive heart failure. Both populations have shown positive responses to IV iron administration (19–21, 46, 50). Most research evaluating the effects of IV iron supplementation on oxidative stress in anemic patients with chronic disease has been conducted on patients undergoing hemodialysis. Overall hemodialysis patients receiving IV iron supplementation have not shown increased rates of mortality or hospitalization from infection. Instead, they have shown improved survival (51). Although hemodialysis patients do show a transient increase in oxidative stress related to IV iron infusions, these have not been shown to have significant clinical implications (51, 56, 57). Patients with congestive heart failure and ACD with functional iron deficiency, who have received IV iron supplementation show improved quality of life, reduced rate of hospitalizations and improved 6-minute walk distance without increased adverse effects (21). Further research is needed to determine if improvements will be seen in all patients with ACD and functional iron deficiency regardless of underlying disease processes.

IV Iron Dosing in COPD

Although there are no evidence-based practice standards available for treating functional iron deficiency in COPD, possible treatment regimens may be extrapolated from data available in other disease states. Currently, there is only one study on IV iron dosing in COPD, performed on a population with chronic renal failure. Since the type of anemia and co-morbidities affecting the patient may impact the overall success of the treatment regimen, further research is needed evaluating which protocols may have the greatest benefit for patients with COPD. Protocols and preparations used in other inflammatory-related anemias may be a good starting point for designing treatment.

When evaluating treatment regimens, the two primary considerations for iron supplementation are the total dose and type of iron to be provided. There are two main techniques for determining treatment dose in iron deficiency anemia. In simple iron deficiency anemia, the total iron deficit is calculated (58). This calculation takes into consideration total estimated blood volume, hemoglobin deficit, and the iron content of hemoglobin. While this method is straightforward, it may not properly account for the tendency to sequester iron in ferritin that is seen in functional iron deficiency. The benefit to this method is that it considers the variation in response that would be seen in patients with significantly different body weights and, therefore, blood volumes. The second method is to provide a standard amount once a patient has been shown to meet designated criteria. Standard dosing regimens are, usually, 1,000–1,500 mg provided in divided doses over a specified period (20, 21, 51). This has been the primary treatment method researched in other inflammatory related iron deficiency states. Patient response is evaluated by rechecking iron parameters upon completion of the dosing regimen. The length of time for repletion is often tied to the IV iron preparation.

Multiple forms of IV iron are available as treatment options, each presenting inherent benefits and limitations (Table 5). Because of these differences, certain iron preparations have not been studied in patients with functional iron deficiencies. The two forms that have been evaluated in the setting of functional iron deficiency are iron sucrose and ferric carboxymaltose (20, 21, 51, 58). Further research within the COPD population is needed to establish best practice standards for both the dose and type of iron, and what parameters or disease outcomes may result with treatment.

Table 5. Comparison of IV iron preparations and their dosing guidelines (51,58–63)

	LMW Iron Dextran*	Ferric Gluconate	Ferumoxytol	Iron Sucrose	Ferric Carboxymaltose
Benefits	Total dose infusion in a single treatment can be accommodated	Well established	Higher dosing threshold	Has been evaluated in studies of functional iron deficiency in CHF	Slower release of iron may provide a lower oxidative burden. Higher dosing threshold.
Limitations	Higher rates of significant adverse drug events Requires a test dose	Only approved for patients receiving dialysis and epoetin therapy	Only approved for adults with iron deficiency and CKD Can alter MRI studies Newer iron preparation	Only approved for patients with chronic kidney disease While single doses up to 500 mg are approved, recommended not to exceed 300 mg	Newer iron preparation.
Product Dosing Guidelines	Daily doses for adults not to exceed 100 mg	125 mg per dialysis session	510 mg dose with a second 510-mg dose 3–8 days later	1,000 mg provided in multiple 200-mg doses over a 14-day period Larger amounts (300 mg) can be accommodated	For patients over 50 kg 1,500 mg provided in two doses of 750 mg at least 7 days apart.

LMW (low molecular weight), CHF (congestive heart failure), CKD (chronic kidney disease), MRI (magnetic resonance imaging).

*Use of high molecular weight iron dextran preparations is not recommended due to higher incidence of adverse reactions.

Implications

Current research has clearly established that anemia is prevalent in a significant portion of the population with COPD and is associated with negative outcomes. Clear evidence exists on the detrimental effects of long-term ACD on iron metabolism. Previous definitions using the criteria for simple iron deficiency will likely prove inadequate. Characteristics utilized in other inflammatory-related disease may better identify iron status when evaluating anemic COPD patients. Future research can clearly define and evaluate the presence of functional iron deficiency within this population.

The role of erythropoietin remains uncertain and may be related to inflammatory processes similar to those that affect functional iron deficiency in ACD (23, 24, 40, 41). It appears that some patients may be resistant to erythropoietin, and it may be argued that ESAs may help to maximize their response to IV iron supplementation. It is also possible that IV iron supplementation alone could be adequate to drive increases in hemoglobin and hematocrit. At this time, there is not enough data to suggest routine use of ESAs in anemic patients with COPD. More studies are needed in this area to better define if ESAs should be considered in anemic patients with COPD and what populations may be considered for treatment.

Although iron supplementation therapy may prove beneficial, the risks must be weighed against the potential benefits for each patient. Iron supplementation is known to cause oxidative stress on the body (46–48), which may be linked to further production of inflammatory cytokines, endothelial dysfunction, and immune deficiency. IV iron provided in excess may lead to increased morbidity and mortality (46, 47, 49, 50). Dosing and frequency regimens studied in other inflammatory-related iron deficiencies are a good starting point in evaluating possible treatment protocols and possibly minimizing some of these potential side effects. Further research will help to define appropriate treatment regimens and limit the chances of excess iron administration.

Finally, more research needs to be conducted on IV iron supplementation in patients with COPD and IDA or ACD with functional iron deficiency. Anemia has been established as an important co-morbidity in COPD yet optimal levels of hemoglobin and hematocrit have yet to be defined. Although data support the negative effects of anemia within these populations, correction of anemia may not necessarily alter patient outcomes. The presence of anemia may only be an indicator of a more advanced disease process. Until more research is done assessing both short- and long-term outcomes, the full effects of reversing anemia will remain unknown. Some studies have demonstrated short-term benefits after improvement in hemoglobin or hematocrit. Unfortunately, these studies were either not conducted with iron supplementation or, in the case of one study, was conducted in a complicated patient population, which limits the generalizability of the results (14, 43–45).

Conclusion

Anemia has been shown in multiple studies to be an important co-morbidity in COPD that is associated with increased morbidity and mortality rates. Lower quality of life, poorer lung function, increased hospital admissions, and both higher health care costs and utilization have each been associated with anemia (2–16). Currently, there is limited evidence to support decisions in treating anemic COPD patients. Iron deficiency plays a role in the development of anemia in both IDA and ACD. Additionally, COPD patients represent a group with tendencies towards inadequate dietary intake of iron and higher risks of deficiency (2, 14, 17, 18, 37).

In patients showing intolerance to oral iron, demonstrating minimal response or decreased intestinal absorption, IV supplementation might need to be considered. Exact cutoffs for acceptable ferritin levels in ACD with functional iron deficiency have yet to be defined. Short- and long-term treatment outcomes in COPD patients have yet to be established. Understanding the underlying causes of anemia in patients with COPD is likely a key defining variable in identifying and evaluating treatment strategies. More research is needed on treating anemia and evaluating possible iron supplementation methods. Information on both long- and short-term results will assist in defining appropriate therapies.

Declaration of Interest Statement

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.