Chronic Obstructive Pulmonary Disease (COPD) is a major cause of morbidity and mortality worldwide.1,2 COPD is associated with various comorbidities such as cardiovascular disease, cachexia, osteoporosis, metabolic syndrome, depression, pulmonary hypertension, lung cancer, and anaemia.1–3 These comorbidities influence mortality and hospitalizations independently, so should be treated appropriately.2,4 Anaemia is increasingly recognized as an important comorbidity,5 being and independent predictor of dyspnoea, exercise limitation, anaerobic threshold, worse health-related quality of life (QoL), and mortality.6–8 Beyond anaemia, non-anaemic iron deficiency (NAID) has also been linked to worse clinical outcomes including frequent exacerbations, and predicts exercise intolerance.6,8

Iron is an essential micronutrient in energy metabolism9 since it plays a key role in oxygen transport (as part of the ferrous ring of haemoglobin (Hb)), storage (as a component of myoglobin), uptake in the oxidative metabolism within the skeletal muscle and in cellular bioenergetics as a whole.10–12 In other words, cellular oxidative metabolism strongly relies on iron availability, which is indispensable for both sufficient oxygen supply, mitochondrial function, and effective substrate catabolism.9 Furthermore, iron is also involved in erythropoiesis.12 Therefore, an adequate iron homeostasis is crucial for cell activity, especially for those characterized by intensive metabolism and high energy demand, such as skeletal myocytes and cardiomyocytes,13 which are key elements in skeletal muscle and heart functions, respectively. In this regard, iron deficiency (ID) may significantly contribute to the loss of skeletal muscle oxidative capacity seen in patients with chronic heart failure (CHF) and COPD.9 Therefore, peripheral and respiratory muscle dysfunction may occur, constituting a potential pathophysiological link between impaired iron status and decreased exercise capacity.14

In recent years, ID with or without anaemia has been studied in pulmonary arterial hypertension (PAH), CHF, and chronic kidney disease, where it was associated with impaired exercise tolerance and worse clinical outcomes.11,13,15 In a previous investigation from our group, COPD patients with NAID showed lower pre-training aerobic capacity and reduced training-induced response in comparison with those with normal iron status, suggesting that ID hampered pulmonary rehabilitation in COPD patients.16 Surprisingly, even though iron supplementation is associated with clinical benefits in patients with PAH or CHF,11,13,15 the impact of this therapy in COPD has not been fully elucidated. In the only randomized clinical trial published to date, intravenous iron supplementation did not show any effect on arterial oxygenation, although it increased exercise capacity improving the six-minute walk test (6MWT).17

The FACE (‘Ferinject Assessment in patients with COPD and iron deficiency to improve Exercise tolerance’) study was designed to test the hypothesis that, compared to placebo, intravenous iron repletion would improve exercise tolerance in patients with ID. Secondary objectives included the assessment of potential changes in QoL and daily physical activity (DPA).

The study flow diagram is detailed in Fig. 2. Although a total of 200 patients were initially assessed for eligibility, only 74 (37.0%) were randomized (50 in the ferric carboxymaltose group and 24 in the placebo group). However, five patients discontinued the trial due to a COPD exacerbation and one was lost in the follow-up. Two more patients who did not replace ID were also excluded. Finally, a total of 66 (33%) patients [44 (66.7%) in the ferric carboxymaltose group and 22 (33.3%) in the placebo group] were included in the analysis. Interestingly, from the consecutive 200 patients that were initially evaluated, 40.5% showed ID.

Fig. 2.

Flow diagram of FACE (Ferinject Assessment in patients with COPD and iron deficiency to improve Exercise tolerance) study according to CONSORT guidelines.

(0.37MB).

Baseline clinical, demographical, and analytical characteristics are described in Table 1. Both groups presented similar baseline characteristics, including in the handgrip and quadriceps strength tests. Median administered ferric carboxymaltose dose was 500 (500–1000) mg.

Improvement of endurance time following iron replacement

As far as 52.3% of patients improved their ET by 33% in the treatment group compared to only a 18.2% in the placebo group (p=0.009; Relative Risk (RR) 3.12, [95% CI, 1.19–8.12], as shown in Fig. 3). The absolute difference was 34.1% (95% CI, 10.20–53.50). Therefore, the treatment of three patients would achieve the primary endpoint in at least 1 patient. Data on the effects of treatment on ET and other clinical outcomes were shown in Table 2. Regarding the change from baseline in ET, statistically significant differences were found between placebo and ferric carboxymaltose group [119.5 (13.8–273.3)s vs. −8.0 (−50.8–69.0)s, p=0.010)]. The inspiratory capacity in the constant work-rate exercise test was assessed in a total of 46 patients (31 in the ferric carboxymaltose group and 15 in the placebo group). No differences were found after the treatment both in patients treated with intravenous iron (p=0.542) or placebo (p=0.875). Handgrip and quadriceps strength tests were also similar after iron treatment between the two study groups: handgrip strength test (iron group p=1.000, placebo group p=0.506) and quadriceps strength test (iron group p=0.793, placebo group p=0.952).

Fig. 3.

Improvement of endurance time following iron replacement among groups.

(0.06MB).

Table 2.

Comparison of main clinical outcomes in the constant work-rate exercise test (CWRET) among ferric carboxymaltose and placebo groups.

	Ferric Carboxymaltose (N=44)	Placebo (N=22)	p value
Endurance time, seconds
Before	279.0 (245.5–344.5)	268.5 (240.8–364.5)	0.927
After	444.0 (304.0–557.5)	252.5 (230.8–385.3)	0.020
Heart rate, bpm
Before	121.5 (108.5–140.5)	117.5 (109.0–128.0)	0.366
Isotime	125.0 (105.0–130.0)	110.5 (105.5–120.0)	0.231
After	128.0 (109.5–138.8)	114.0 (103.0–127.5)	0.017
Minute ventilation, L/min
Before	38.9 (30.2–46.3)	40.7 (32.5–49.5)	0.546
After	39.3 (31.8–52.3)	40.1 (32.3–49.0)	0.483
Final dyspnoea (Borg score)
Before	5.0 (3.0–7.0)	6.0 (4.5–7.5)	0.238
After	4.0 (3.0–7.0)	5.5 (4.0–7.0)	0.187
Final leg discomfort (Borg score)
Before	7.0 (5.0–8.0)	6.0 (4.0–7.0)	0.308
After	5.0 (4.0–8.0)	6.5 (4.8–8.3)	0.118

Data are presented as median (Q1–Q3).

Bpm: beats per minute.

From the 44 patients treated with intravenous iron, 4 (9.1%) presented mild anaemia. These patients ameliorated their ET by 238.0 (78.8–264.5) s, p=0.043, all of them achieving the primary endpoint. No differences were found in the change from baseline in ET compared to patients with NAID [238.0 (78.8–264.5) vs. 103.5 (9.3–278.0), p=0.541] or Hb [1.2 (−0.1–1.8) vs. 0.5 (−0.2–1.2) g/dL, p=0.383].

To the best of our knowledge, this is the largest study assessing the impact of the intravenous iron supplementation on exercise tolerance in COPD patients with ID with or without mild anaemia. The main finding of this study was that the treatment with intravenous ferric carboxymaltose was associated with a larger improvement in the endurance time compared to placebo, even in patients without mild anaemia. The results of our study confirm our previous hypothesis and fit with those found in other chronic diseases. Based on our predefined subgroup analysis, we also found that patients with a ferritin value<30ng/mL performed strikingly better. However, this is only a hypothesis generating data. Finally, iron treatment was well tolerated without significant safety issues.

ID is estimated to affect two billion people worldwide23 being the leading cause of anaemia.24 Moreover, ID appears to be responsible for 0.8% of all deaths and 1.3% of all disability-adjusted life years.24 In COPD patients in particular, anaemia and ID are quite common. However, the former is neither sought for or, if found, not treated.25 Furthermore, NAID is observed in almost 40–50% of patients.8,26 The present study shows a NAID prevalence of 40% in a screening of 200 patients. However, although ID and NAID are associated with reduced exercise tolerance and QoL,16 current guidelines of COPD2 did not recognize these abnormalities as common and clinically relevant comorbidities. In contrast, the 2016 European Society of Cardiology guidelines for the diagnosis and treatment of acute and CHF do recommend ferric carboxymaltose for the treatment of ID to alleviate heart failure symptoms and improve exercise capacity and QoL.27,28

To date, studies on ID and resultant decreased exercise performance have been performed predominantly in animal models, while data regarding the relationship between ID and skeletal muscle dysfunction in ID anaemic human subjects are limited and inconsistent.9 Furthermore, the benefits of iron replacement have been found in cardiopulmonary diseases other than COPD such as CHF and PAH with ID, with or without anaemia.11,13,15,29–32 In the largest randomized controlled clinical trial including 459 patients with CHF, Anker et al.11 showed that treatment with intravenous ferric carboxymaltose improved symptomatology, functional capacity and QoL with an acceptable safety profile. Similarly, this treatment can also improve clinical outcomes in PAH patients with ID.33 Moreover, Viethen et al.32 studied the treatment with a single dose of ≤1000mg of ferric carboxymaltose in 20 patients with pulmonary hypertension and ID, reporting a significant improvement not only in iron status but in the 6MWT distance and QoL.

While the effort limitation mechanisms in iron deficiency anaemia may also include oxygen transport, when there is NAID, it might be the result of diminished oxygen utilization in the muscle cells due to impaired synthesis or function of different iron-containing enzymes.6 In fact, experimental data from in vitro and animal studies indicate that skeletal muscle network of pathways for energy production is hampered by ID at different points in the distinct metabolic routes.9 ID affects mitochondrial morphology (such as a decrease in density of cristae in the mitochondrial inner membrane) and dramatically impairs the process of mitochondrial oxidative phosphorylation.9 All these effects may add to the general decrease in oxidative metabolism efficiency.9 This is one of the hypotheses by which NAID may have a negative impact on aerobic capacity16 and can produce fatigue and exercise intolerance34 in patients with demonstrated peripheral muscle dysfunction.9,35 On the other hand, it should be mentioned that peripheral muscle strength did not improve following treatment with iron in any of the study groups. These results suggest that factors other than muscle function are major contributors to general exercise tolerance in severe COPD patients, as has been shown in previous reports.36–39 As such, it has been demonstrated a mismatch between ventilatory capacity and the patients’ demand, an imbalance between energy requirements and supply to the working respiratory and peripheral muscles, and other factors that contribute to muscle dysfunction.37 The effects of iron supplementation may go beyond those of skeletal muscle performance. Organs such as the vascular endothelium, myocardium, and the lungs and airways may have also contributed to the improvement in ET seen in the patients who received iron replacement.36–39 Another interesting possibility is the decrease in myoglobin reserves in the muscle, which would affect its bioenergetics. As Hb, myoglobin is an iron- and -oxygen binding protein. Its function is not completely elucidated but it is believed that it also participates in oxygen storage within the muscle increasing the availability of this gas for aerobic purposes.40 On the other hand, myoglobin as Hb are scavengers of reactive oxygen species.41 This would limit oxidative stress, one of the mechanisms involved in muscle dysfunction repeatedly observed in COPD.42

Considering all this evidence, it is not surprising that in the aforementioned randomized clinical trial conducted in patients with CHF, the treatment with ferric carboxymaltose was beneficial to both patients with anaemia and those without anaemia.11 Our study clearly shows similar results in moderate to severe COPD patients either with mild anaemia or NAID. This suggests that the presence of ID is a valid independent therapeutic target in different chronic diseases. We hypothesize that iron availability in skeletal muscles could have improved their oxidative capacity,14 increasing the time to the anaerobic threshold during submaximal exercise.31 The only previous clinical trial analysing the impact of intravenous iron in COPD was recently published by Santer et al.17 Their primary endpoint (oxygen saturation) was not met although exercise capacity (assessed by the 6MWT) improved. However, this might be explained by the fact that patients’ baseline oxygen saturation was only marginally (<1%) lower than that of age-matched general population.17 The exact mechanisms leading to anaemia and ID in COPD are complex and remain not fully understood. Moreover, iron metabolism in patients with this and other chronic illness merits a more detailed investigation to better unravel the reasons why the correction of ID can result in symptomatic improvements even in the absence of changes in Hb.

Interestingly, we have also observed that the improvement in exercise capacity was especially important in those patients with ferritin<30mg/dL.24,43 This is probably related to the main functions of this protein, which are iron storage, transportation, and delivery to the tissues. However, our finding deserves further studies to completely understand the mechanisms of such a higher improvement.

Our study achieves the primary endpoint showing a 3-fold possibility of a 33% improvement in the endurance time with iron treatment when compared to placebo administration. Although this specific improvement (33%) was defined by the European Respiratory Society task force for nonpharmacological interventions,44 we strongly believe that it should be also valid for pharmacological treatments. On the other hand, the election of the endurance time during CWRET as the primary endpoint was based on the previous use of this variable as the outcome measure in other studies, including those using ferric carboxymaltose.19,45

Regarding the secondary endpoints of the present study, we also observed an improvement in CAT questionnaire in those patients who received ferric carboxymaltose. However, DPA did not change. Our hypothesis for this latter result is that all patients were sedentary, and changes in this dimension normally also involve a behavioural intervention.46

The FACE study has shown that intravenous iron carboxymaltose therapy is well tolerated and associated with improvements in exercise tolerance and symptoms in patients with COPD and iron deficiency, with or without anaemia. These benefits were even more evident in patients with low ferritin levels. The assessment of the iron profile and correction of iron status may therefore become part of the follow-up and management of COPD patients. Further studies will answer several unsolved questions such as the potential underlying molecular mechanisms, the long-term benefits of the treatment, and the needs for more infusions.

This study has been funded by Instituto de Salud Carlos-III (ISCIII), contract grants FIS 17/00649 and 18/00075 [FEDER]), BA 17/00025. CIBERES, ESF02/2017; Spanish Respiratory Society (SEPAR) 409/2017; Catalan Foundation of Pulmonology (FUCAP) 2017; and an unrestricted grant from Vifor Pharma 2018. Vifor Pharma did not have any role in the study or statistical analyses.

None.