Summary
Background
Functional status is one of the main concerns in the management of heart failure (HF). Recently, the FAIR-HF and CONFIRM-HF trials showed that correcting anaemia using intravenous iron supplementation improved functional variables in patients with absolute or relative iron deficiency. Relative iron deficiency is supposed to be a marker of HF severity, as ferritin concentration increases with advanced stages of HF, but little is known about the impact of absolute iron deficiency (AID).
Aims
To study the impact of AID on functional variables and survival in patients with chronic systolic HF.
Methods
One hundred and thirty-eight non-anaemic patients with chronic systolic HF were included retrospectively. Patients were divided into two groups according to iron status: the AID group, defined by a ferritin concentration < 100 μg/L and the non-AID group, defined by a ferritin concentration ≥ 100 μg/L. Functional, morphological and biological variables were collected, and survival was assessed.
Results
Patients in the AID group had a poorer 6-minute walking test (342 vs. 387 m; P = 0.03) and poorer peak exercise oxygen consumption (13.8 vs. 16.0 mL/min/kg; P = 0.01). By multivariable analysis, ferritin < 100 μg/L was associated with impaired capacity of effort, assessed by peak exercise oxygen consumption. By multivariable analysis, there was no difference in total mortality between groups, with a mean follow-up of 5.1 ± 1.1 years.
Conclusions
The poorer functional evaluations in iron-deficient patients previously reported are not caused by the merging of two different populations (i.e. patients with absolute or relative iron deficiency). Our study has confirmed that non-anaemic HF patients with AID have poorer peak oxygen consumption. However, AID has no impact on the survival of these patients.
Résumé
Contexte
Le statut fonctionnel du patient insuffisant cardiaque constitue un élément majeur de la qualité de vie et son amélioration un des buts du traitement. Récemment, les essais contrôlés FAIR-HF et CONFIRM-HF ont démontré que le fer intraveineux peut améliorer l’état fonctionnel des patients présentant une carence martiale absolue ou fonctionnelle. Cependant, à la différence d’une carence martiale absolue, une carence relative en fer ne pourrait être qu’un simple marqueur de la sévérité de l’insuffisance cardiaque, les taux de ferritine augmentant dans les stades avancés de l’insuffisance cardiaque en raison d’un état inflammatoire chronique.
Objectifs
Le but de cette étude est de déterminer si les patients insuffisants cardiaques non anémiques présentant une carence absolue en fer ont un statut fonctionnel et une survie différents des patients non carencés en fer.
Méthodes
Dans cette étude rétrospective, 138 patients insuffisants cardiaques avec altération de la fonction systolique ventriculaire gauche (FE < 40 %) en stade 2 et 3 de la New York Heart Association, sans anémie, ont été inclus et séparés en 2 groupes en fonction de la valeur de la ferritinémie, inférieure ou supérieure à 100 μg/L. Une évaluation fonctionnelle, morphologique et biologique a été effectuée. Une analyse de la survie en fonction de l’existence ou non d’une carence martiale absolue a été réalisée.
Résultats
Les patients avec une carence martiale absolue (ferritinémie < 100 μg/L) présentaient une distance au test de marche de 6 minutes (342 contre 387 m ; p = 0,03) et une consommation maximale d’oxygène à l’effort (13,8 contre 16,0 mL/min/kg ; p = 0,01) moindres que les patients sans carence absolue en fer (ferritinémie > 100 μg/L). En analyse multivariée, la ferritine inférieure à 100 μg/L était associée à une altération de la capacité à l’effort, évaluée par VO 2 max. En analyse univariée, aucune différence n’a été objectivée pour le risque de décès entre les deux groupes après un suivi moyen de 5.1 ± 1.1 ans.
Conclusions
La diminution des performances fonctionnelles précédemment observée chez les patients carencés en fer n’est pas liée à la superposition de deux populations différentes : les patients présentant une carence martiale absolue ou fonctionnelle. Ces données confirment que les patients insuffisants cardiaques avec une carence martiale absolue ont une capacité maximale de consommation d’oxygène plus faible. Cependant, l’existence d’une carence martiale absolue n’influence pas la survie de ces patients.
Background
Improvement of dyspnoea and functional status still remains one of the main concerns in the treatment of heart failure (HF). It is now known that anaemia is a strong worsening factor for both functional status and survival in HF ; its prevalence in HF ranges from 23% to 57% . Paradoxically, correction of anaemia by erythropoietin did not improve the prognosis of patients with HF in the RED-HF trial . Although anaemia in HF can be multifactorial, iron deficiency remains the main cause and was shown recently to be a strong predictive factor for death and morbidity in a population that included anaemic and non-anaemic patients . Numerous studies have demonstrated that correcting anaemia using intravenous iron administration leads to an improvement in functional variables (such as the Minnesota Living with Heart Failure Questionnaire, New York Heart Association [NYHA] stage or 6-minute walking test), renal function, B-type natriuretic peptide (BNP) concentration and left ventricular ejection fraction (LVEF) . The recent FAIR-HF and CONFIRM-HF trials showed that intravenous iron supplementation by ferric carboxymaltose led to an improvement in functional variables in patients with iron deficiency, independent of anaemia. These results suggest that prognosis and functional status in HF are linked more to the iron deficiency than to the anaemia itself. Diagnosis of iron deficiency usually considers two different populations: patients with an absolute iron deficiency (AID), defined by a low ferritin concentration, indicating low iron reserve; and patients with a relative iron deficiency, defined by a normal or high ferritin concentration, but a low transferrin saturation rate (TSR). In this last group, low TSR is the result of iron sequestration secondary to an increase in haptoglobin, and despite a high production of ferritin induced by inflammation. Consequently, because ferritin concentration increases with inflammation in patients with advanced HF , we can suppose that the poorer functional status and survival of patients with relative iron deficiency are the result of a more advanced disease. The physiopathology of AID is very different and results from a decrease in iron stores.
The aim of our study was to investigate the impact of AID, independent of anaemia, on the functional status and survival of patients with chronic systolic HF.
Methods
Population
Patients referred to our tertiary centre for HF at the University Hospital of Toulouse for optimisation of their therapy, implantation of a cardioverter defibrillator, multisite pacing, or evaluation of heart transplantation or assistance device indication were included retrospectively between January 2007 and December 2008. All patients had undergone coronary angiography within the past 2 years, and the diagnosis of ischaemic cardiomyopathy was retained when there was > 50% stenosis of at least one principal coronary artery. Patients with decompensated HF, defined by the exacerbation of typical symptoms of HF within the last month, or anaemia, defined by a haemoglobin concentration < 13.0 g/dL in men or < 12.0 g/dL in women, were excluded from the study. Patients without anaemia were separated into two groups: patients with AID, defined by a ferritin concentration < 100 μg/L and patients without AID, defined by a ferritin concentration ≥ 100 μg/L.
Clinical testing
Functional variables (including NYHA stage, 6-minute walking test and peak oxygen consumption [peak VO 2 ]), LVEF and BNP were compared between groups. Peak VO 2 was assessed with a cycling ergometer; the protocol started at 20 W, with an increase of 10 W each minute, and consisted of measuring the maximal oxygen uptake in last 30 seconds of a maximal effort. The 6-minute walking test consisted of measuring the maximal distance covered by the patient in 6 minutes. LVEF was determined by transthoracic echocardiography, using the conventional apical two- and four-chamber views and the modified Simpson’s biplane method.
Follow-up
Clinical follow-up was done by a telephone interview with the patient’s general practitioner or cardiologist, with the patient or with their family. The outcome event examined was all-cause mortality. Patients who underwent cardiac transplantation during follow-up were considered as dead at the date of intervention. Patients without contact for up to 6 months were considered as lost to follow-up and were excluded from the survival analysis.
Statistical analysis
Continuous variables were tested for normal distribution using the Kolmogorov–Smirnov test and are expressed as means ± standard deviations. Categorical data are expressed as numbers and percentages. Age, BNP, haemoglobin, creatinine and LVEF were considered in tertiles because of a lack of linearity. Ferritin was analysed as a dichotomous variable (< 100 μg/L or ≥ 100 μg/L). The association between categorical variables was investigated by the χ 2 test and the mean values of continuous variables were compared using Student’s t test. Correlation between variables was assessed using Spearman’s test or Pearson’s test when required, and is expressed as r . All-cause mortality was summarized using Kaplan–Meier survival curves, and the log-rank test was used for initial comparisons. A stepwise selection was done using a P to remove from and a P to enter into the model of 0.05, with previous backward selection after inclusion of all selected variables (saturated model) and then forward selection. The P value refers to the likelihood ratio test of the hypothesis that the regression coefficient was zero. Results are expressed as relative risks with 95% confidence intervals. A significant increase in risk is obtained if the 95% confidence interval excludes 1 and the Wald test P value is < 0.05. Differences were considered statistically significant for P values < 0.05. All analyses were performed using StatView, version 5 (SAS Institute Inc., Cary, NC, USA).
Results
Population characteristics
Among the 202 patients screened, 64 (31.6%) patients had anaemia and were excluded from the analysis. Among the 138 patients included, the mean age was 60.9 ± 14.5 years, 69 (50%) were men, and the rate of ischaemic aetiology was low (35.2%). Population characteristics according to iron deficiency are presented in Table 1 . As already described in the general population, there were more women in the AID group than in the non-AID group (64% vs. 26%; P = 0.01). Sex was strongly correlated with iron deficiency (R = 0.64; P < 0.001); for the total population of non-anaemic patients, AID was found in 36 (52%) women versus 10 (15) men ( P < 0.001). The mean ferritin concentration was 487 ± 714 μg/L for men and 113 ± 66 μg/L for women ( P < 0.001).
