Iron Deficiency Key Determinant of HRQoL in HF
Iron Deficiency Key Determinant of HRQoL in HF
In this study, we have shown that HRQoL reported by CHF patients in a multidisciplinary heart failure programme was adversely affected by ID (ferritin/TSAT or sTfR levels) independent of anaemia, and that the impairment was mostly due to elements of the MLHFQ physical domain rather than the emotional domain. Furthermore, although anaemia and Hb levels were associated with HRQoL in univariate analyses (Figure 1), in multivariable models (Figure 2, Table 3) they had no statistically significant effect on HRQoL when the effects of ID and other variables were taken into account. Interestingly, anaemia in the presence of ID was associated with greater iron depletion than in ID alone.
Iron is physiologically important for erythropoiesis and oxygen transport, and, as such, ID is one of the main causes of anaemia. It also plays a vital role in muscle function and exercise capacity that is independent of anaemia. Specifically in the context of CHF, there is accumulating evidence that ID may contribute to myocardial alterations, to fatigue and reduced exercise capacity, and to increased risk of mortality. In clinical trials, i.v. iron treatment can improve CHF symptoms, increase exercise capacity, and improve HRQoL in patients with ID with or without anaemia. These aspects have been confirmed in a recent meta-analysis. We observed that in patients with ID, sTfR concentrations were higher (indicating more pronounced ID) in anaemic compared with non-anaemic patients and that ID but not anaemia was independently associated with HRQoL. From this, we could hypothesize that ID is the key factor affecting self-perceived health status, and anaemia would merely be a marker of more pronounced iron depletion. Hypothetically, ID would impair submaximal exercise capacity, which in turn affects the performance of usual activities, a crucial determining factor in HRQoL. Further studies are needed to confirm this hypothesis.
Chronic heart failure severity itself is known to correlate with HRQoL. In multivariate analyses, we observed that a number of clinical variables were independently associated with impaired HRQoL, including CHF-related factors (NYHA class, time since last hospital admission for CHF, and administration of loop diuretics), as well as systolic blood pressure, presence of diabetes, renal function, and level of C-reactive protein (Table 3). However, neither LVEF, anaemia (Hb ≤ 12 g/dL), nor Hb as a continuous variable were independent predictors of impaired HRQoL, although anaemia might be associated with increased iron depletion in patients with ID. Interestingly, the effect of ID on HRQoL was also independent of the level of functional limitation determined with the NYHA class. This underscores the multidimensional nature of HRQoL since other factors besides dyspnoea, such as fatigue or pain, may entail a limitation of submaximal exercise capacity which in turn would interfere with usual activities, affecting the self-perceived health status of chronic patients. Regarding this, data coming from the FAIR-HF study show that i.v. iron repletion improved the domain of the EQ5D measuring the impact on pain and discomfort of patients with ID and CHF compared with placebo. Further study is required to understand the implications of these associations and their role in CHF management.
Despite the negative impact of ID on CHF, there is no currently accepted definition of ID in this area. Therefore, we took the definition of the nephrology K/DOQI guidelines (ferritin levels <100 ng/mL, or < 800 ng/mL with TSAT < 20%) as a starting point, but we acknowledge that the choice of markers and cut-offs in nephrology might not be suitable in CHF. For example, the presence of inflammation may disturb iron metabolism and complicate the interpretation of traditional markers such as ferritin or TSAT. Therefore, we considered measurement of sTfR, the truncated fragment of the membrane receptor of transferrin, which is unaffected by inflammation or gender. Several studies using bone marrow examination as a gold standard for iron status evaluation have shown that the sTfR level is an accurate measure of iron demand in patients with or without anaemia. It also offers the opportunity to evaluate iron levels as a continuum from normal through to absolute depletion, which might accurately reflect the pathophysiological mechanisms involved in the development of abnormal iron status in patients with chronic diseases such as CHF. Some authors suggest that the ferritin index (sTfR/log10[ferritin]) would also be an ideal marker for evaluating iron metabolism in patients with chronic conditions. This index combines information about iron demand (sTfR) and availability of iron (log10[ferritin]) for erythropoiesis and for other functional proteins. In our study, we focused on sTfR and, thus, further studies of the role of the ferritin index in the evaluation of iron metabolism in CHF are warranted.
We found that impaired HRQoL, adjusted for the effects of other variables identified in univariate analyses, correlated strongly with elevated sTfR when expressed in quintiles (P < 0.001). On the other hand, a similar analysis for Hb levels revealed no statistically significant relationship with HRQoL. This was expected given the absence of an association between anaemia and HRQoL in this cohort. We therefore recommend further study to assess the suitability of sTfR as a marker of ID in CHF, given the potential of this biomarker to express iron status as a spectrum of severity of iron depletion rather than a dichotomous state as in the K/DOQI definition.
These results add to the emerging understanding of the role of ID in CHF, and may help explain the results of clinical trials showing the benefits of i.v. iron treatment in CHF patients. However, the study is subject to some limitations. This was a post-hoc analysis of data from a single centre, so it is unclear how applicable the findings are to other countries and ethnicities, although a study in Poland has also shown the negative impact of ID on CHF independently of anaemia. This was a cross-sectional study, thus, assumptions about a causative role of ID in impaired HRQoL can only be hypothesized. However, interventional studies showing improvement of HRQoL with i.v iron in patients with ID and CHF support this hypothesis. Patients excluded from this analysis differ from the analysed cohort in terms of age, LVEF, and NYHA class. We cannot ascertain the effect of including the 253 excluded patients on the final results of this analysis. However, in our multivariable analyses, the influence of ID on HRQoL was independent of these factors and thus we may hypothesize that the results would not significantly differ from those presented. Given these limitations, and the lack of power to detect differences between groups, prospective, appropriately powered, multicentre studies are necessary to confirm the results. Although specific questionnaires, such as the MLHFQ, are superior to generic questionnaires to describe self-perceived health status in CHF patients, generic questionnaires allow comparisons between chronic conditions and provide additional information about important aspects in CHF that may be missed using specific tools. Thus, it would be desirable to include these generic tools in future research. Despite the limitations mentioned here, we have studied typical CHF patients, who are not always represented in clinical trials. Therefore, our observations may more closely reflect the real world than a controlled trial with rigorous inclusion and exclusion criteria. Finally, the analysis represents a single time point in the lives of these patients, and as such does not capture dynamic effects of the disease or the impact of treatment, including i.v. iron. Future studies would be desirable to analyse the changes of indices of iron status over time and the impact of these changes on measures of HRQoL.
In conclusion, we have shown that CHF patients with ID suffer impaired HRQoL and that the effects are not due to the role of iron in erythropoiesis. In patients with ID, anaemia itself had no effect on HRQoL, though reduced Hb levels could be associated with increased levels of iron depletion. Further study is required to deepen our understanding of the role iron plays in CHF aetiology and severity.
Discussion
In this study, we have shown that HRQoL reported by CHF patients in a multidisciplinary heart failure programme was adversely affected by ID (ferritin/TSAT or sTfR levels) independent of anaemia, and that the impairment was mostly due to elements of the MLHFQ physical domain rather than the emotional domain. Furthermore, although anaemia and Hb levels were associated with HRQoL in univariate analyses (Figure 1), in multivariable models (Figure 2, Table 3) they had no statistically significant effect on HRQoL when the effects of ID and other variables were taken into account. Interestingly, anaemia in the presence of ID was associated with greater iron depletion than in ID alone.
Iron is physiologically important for erythropoiesis and oxygen transport, and, as such, ID is one of the main causes of anaemia. It also plays a vital role in muscle function and exercise capacity that is independent of anaemia. Specifically in the context of CHF, there is accumulating evidence that ID may contribute to myocardial alterations, to fatigue and reduced exercise capacity, and to increased risk of mortality. In clinical trials, i.v. iron treatment can improve CHF symptoms, increase exercise capacity, and improve HRQoL in patients with ID with or without anaemia. These aspects have been confirmed in a recent meta-analysis. We observed that in patients with ID, sTfR concentrations were higher (indicating more pronounced ID) in anaemic compared with non-anaemic patients and that ID but not anaemia was independently associated with HRQoL. From this, we could hypothesize that ID is the key factor affecting self-perceived health status, and anaemia would merely be a marker of more pronounced iron depletion. Hypothetically, ID would impair submaximal exercise capacity, which in turn affects the performance of usual activities, a crucial determining factor in HRQoL. Further studies are needed to confirm this hypothesis.
Chronic heart failure severity itself is known to correlate with HRQoL. In multivariate analyses, we observed that a number of clinical variables were independently associated with impaired HRQoL, including CHF-related factors (NYHA class, time since last hospital admission for CHF, and administration of loop diuretics), as well as systolic blood pressure, presence of diabetes, renal function, and level of C-reactive protein (Table 3). However, neither LVEF, anaemia (Hb ≤ 12 g/dL), nor Hb as a continuous variable were independent predictors of impaired HRQoL, although anaemia might be associated with increased iron depletion in patients with ID. Interestingly, the effect of ID on HRQoL was also independent of the level of functional limitation determined with the NYHA class. This underscores the multidimensional nature of HRQoL since other factors besides dyspnoea, such as fatigue or pain, may entail a limitation of submaximal exercise capacity which in turn would interfere with usual activities, affecting the self-perceived health status of chronic patients. Regarding this, data coming from the FAIR-HF study show that i.v. iron repletion improved the domain of the EQ5D measuring the impact on pain and discomfort of patients with ID and CHF compared with placebo. Further study is required to understand the implications of these associations and their role in CHF management.
Despite the negative impact of ID on CHF, there is no currently accepted definition of ID in this area. Therefore, we took the definition of the nephrology K/DOQI guidelines (ferritin levels <100 ng/mL, or < 800 ng/mL with TSAT < 20%) as a starting point, but we acknowledge that the choice of markers and cut-offs in nephrology might not be suitable in CHF. For example, the presence of inflammation may disturb iron metabolism and complicate the interpretation of traditional markers such as ferritin or TSAT. Therefore, we considered measurement of sTfR, the truncated fragment of the membrane receptor of transferrin, which is unaffected by inflammation or gender. Several studies using bone marrow examination as a gold standard for iron status evaluation have shown that the sTfR level is an accurate measure of iron demand in patients with or without anaemia. It also offers the opportunity to evaluate iron levels as a continuum from normal through to absolute depletion, which might accurately reflect the pathophysiological mechanisms involved in the development of abnormal iron status in patients with chronic diseases such as CHF. Some authors suggest that the ferritin index (sTfR/log10[ferritin]) would also be an ideal marker for evaluating iron metabolism in patients with chronic conditions. This index combines information about iron demand (sTfR) and availability of iron (log10[ferritin]) for erythropoiesis and for other functional proteins. In our study, we focused on sTfR and, thus, further studies of the role of the ferritin index in the evaluation of iron metabolism in CHF are warranted.
We found that impaired HRQoL, adjusted for the effects of other variables identified in univariate analyses, correlated strongly with elevated sTfR when expressed in quintiles (P < 0.001). On the other hand, a similar analysis for Hb levels revealed no statistically significant relationship with HRQoL. This was expected given the absence of an association between anaemia and HRQoL in this cohort. We therefore recommend further study to assess the suitability of sTfR as a marker of ID in CHF, given the potential of this biomarker to express iron status as a spectrum of severity of iron depletion rather than a dichotomous state as in the K/DOQI definition.
These results add to the emerging understanding of the role of ID in CHF, and may help explain the results of clinical trials showing the benefits of i.v. iron treatment in CHF patients. However, the study is subject to some limitations. This was a post-hoc analysis of data from a single centre, so it is unclear how applicable the findings are to other countries and ethnicities, although a study in Poland has also shown the negative impact of ID on CHF independently of anaemia. This was a cross-sectional study, thus, assumptions about a causative role of ID in impaired HRQoL can only be hypothesized. However, interventional studies showing improvement of HRQoL with i.v iron in patients with ID and CHF support this hypothesis. Patients excluded from this analysis differ from the analysed cohort in terms of age, LVEF, and NYHA class. We cannot ascertain the effect of including the 253 excluded patients on the final results of this analysis. However, in our multivariable analyses, the influence of ID on HRQoL was independent of these factors and thus we may hypothesize that the results would not significantly differ from those presented. Given these limitations, and the lack of power to detect differences between groups, prospective, appropriately powered, multicentre studies are necessary to confirm the results. Although specific questionnaires, such as the MLHFQ, are superior to generic questionnaires to describe self-perceived health status in CHF patients, generic questionnaires allow comparisons between chronic conditions and provide additional information about important aspects in CHF that may be missed using specific tools. Thus, it would be desirable to include these generic tools in future research. Despite the limitations mentioned here, we have studied typical CHF patients, who are not always represented in clinical trials. Therefore, our observations may more closely reflect the real world than a controlled trial with rigorous inclusion and exclusion criteria. Finally, the analysis represents a single time point in the lives of these patients, and as such does not capture dynamic effects of the disease or the impact of treatment, including i.v. iron. Future studies would be desirable to analyse the changes of indices of iron status over time and the impact of these changes on measures of HRQoL.
In conclusion, we have shown that CHF patients with ID suffer impaired HRQoL and that the effects are not due to the role of iron in erythropoiesis. In patients with ID, anaemia itself had no effect on HRQoL, though reduced Hb levels could be associated with increased levels of iron depletion. Further study is required to deepen our understanding of the role iron plays in CHF aetiology and severity.
Source...