Hemoglobin Levels and Health-related Quality of Life
Almost all of the papers dealing with chronic kidney disease (CKD) invariably start underlining that morbidity and mortality rate of this patient population is unacceptably high. Conversely, very few papers have considered quality of life and none selected quality of life as the primary end point. This reflects a clear bias for doctors and investigators, possibly because the treatment of CKD, especially in stage 5 on dialysis or with transplantation, is life-saving. Thus, the efforts of nephrologists were and are still concentrated in reducing the mortality and the morbidity of patients, without carefully considering quality of life. This attitude has been amplified by the influence of evidence-based medicine and regulatory authorities: the only end points that really matter are the hard-ones. Surrogate end points, especially when they are not easily quantifiable, are considered often insufficient to prove the efficacy of a given treatment. This process has markedly improved the scientific quality of clinical trials. However, we are reaching the paradox that no clinical trial is capable of proving efficacy on hard end points. Recently, almost all large, well-designed clinical trials testing a variety of treatments have given negative results. This has led physicians back to the importance of individualizing the cure of the single patient according to his characteristics and it is revaluating a comprehensive view of the possible associated benefits of a given treatment.
Apart from treating hypertension and dyslipidaemia, anaemia correction is an important potential area for reducing cardiovascular morbidity and mortality in CKD.
Before the availability of erythropoiesis stimulating agents (ESA), CKD patients, especially those on dialysis, were heavily anaemic and symptomatic. Blood transfusions partially and temporary alleviated symptoms, but at the price of the risk of viral infections, iron overload and sensitization, which could jeopardize the possibility of receiving a transplant.
Since the early 1990s, the introduction of ESA has revolutionized the quality of life of CKD patients, who previously had very limited possibility not only to maintain a job, but also to move: they could barely survive and did not enjoy life. Doctors very soon realized the potential of ESAs. These agents were very effective in increasing haemoglobin (Hb) levels even to normal values. A striking improvement in quality of life was the second most evident effect. However, despite this marked improvement in patient well-being, in 1989 the Food and Drug Administration (FDA) approved Epoetin alfa 'to elevate or maintain the red blood cell level and to decrease the need for transfusions', without even mentioning quality of life.
Observational data from large databases have invariably shown a clear relationship between anaemia and outcomes: the worse the anaemic status, the worse the patient mortality and morbidity. The possibility of improving survival and reducing comorbidities through anaemia correction became the aim of several clinical trials. Perhaps forgetting that mild anaemia may have a compensatory role in CKD patients, trials were designed to compare near-to-normal Hb target ranges with partial anaemia correction.
Back to the 1990s, the Normal Hematocrit study was stopped for futility, together with trend towards a greater risk of death or myocardial infarction in the patient group that was randomized to achieve a normal haematocrit.
The general feeling of nephrologists was that ESA treatment was possibly started too late, when the cardiovascular damage was too advanced to obtain benefit with anaemia treatment. However, subsequent trials, which enrolled non-dialysis, less compromised patients, gave disappointing results as well.
In 2009, the publication of the Trial to Reduce Cardiovascular Events with Aranesp® Therapy (TREAT) sanctioned the end of the dream of improving survival and the risk of cardiovascular events by means of complete anaemia correction. Despite neutral findings at the primary analysis, secondary analyses showed a higher risk of stroke and death for cancer in patients with a history of strokes and malignancies, respectively.
The wake of these findings, further strengthened by the publication of Kidney Disease: Improving Global Outcomes (KDIGO) guidelines (and possibly by changes in ESA reimbursement practice in the USA), has led to a progressive reduction of ESA doses and achieved Hb values. Today, an increasing percentage of CKD patients experience severe anaemia and suffer those anaemia-related symptoms that were only observed in the pre-ESA era and always more frequently receive blood transfusions.
In this uncertain scenario, quality of life has to be reconsidered and weighted with the risk of adverse events in the single patient.
According to the common experience of physicians in everyday clinical practice, early trials using erythropoietin showed improvements in multiple domains and measures of quality of life. In particular, data of three of these early trials were put together to assess the effect of erythropoietin treatment on quality of life in over 1500 patients. Haemodialysis patients treated with rHuEPO had a significant improvement in objective (functional ability and health status) and subjective (well-being, life satisfaction, psychological aspects and happiness) quality of life indicators at 10 months. A newer analysis of the Canadian Erythropoietin Study also showed a significant increase in exercise tolerance and patient-reported physical function in the epoetin-alfa treated patients compared with those receiving placebo. Improvements were seen at 2 months and were maintained through Months 4 and 6.
The interpretation of these studies was complicated by methodological limitations (no control group) and by the fact that instruments for measuring quality of life and achieved Hb levels were heterogeneous. The median follow-up was quite short, not giving information whether changes in quality of life are long lasting. Indeed, these data were reviewed by the FDA and considered not convincing.
Several meta-analyses tried to summarize the effect of anaemia correction on quality of life in both dialysis and non-dialysis patients and showed a consistent and positive impact on several aspects of quality of life. However, they inevitably suffered the methodological limitations of the included studies.
Recent clinical trials used more standardized measures of quality of life. The Cardiovascular Risk Reduction by Early Anaemia Treatment (CREATE) study, which was a large, open-label study, only tested quality of life with the Short-Form Health Survey (SF-36) questionnaire. At 1 year, anaemia correction towards Hb levels >13 g/dL was associated with a moderate effect on quality of life compared with partial correction (Hb 10.5–11.5 g/dL) with regard to general and mental health, physical function, physical role, social function and vitality. At the end of the second year, the significant difference between groups was maintained only for general health and vitality. The Correction of Haemoglobin and Outcomes in Renal Insufficiency (CHOIR) study used a larger set of instruments to test quality of life [Linear Analogue Self-Assessment (LASA), the Kidney Disease Questionnaire (KDQ) and SF-36]. Both patients in the low- and high Hb group reported an improvement in energy and physical function score without any difference between the two groups. A higher burden of comorbidities and a small difference in the mean achieved Hb levels between the two groups (1.3 g/dL) may possibly explain why these findings are not concordant with those of the CREATE trial.
The TREAT is the largest double-blind, placebo controlled trial in the field. In this study, QOL was measured with Functional Assessment of Cancer Therapy (FACT)-Fatigue, Short Form-36, and EuroQol scores over a follow-up period of nearly 2 years. Overall, health-related quality of life (HRQOL) improved more in the darbepoetin compared with the placebo group. However, at both the adjusted and unadjusted analysis, only the FACT-fatigue score was significantly improved in the patients treated with darbepoetin compared with placebo. This effect was judged of modest clinical significance. Of note, a higher percentage of patients in this group had a ≥3-point increase (i.e. a clinically meaningful change) of the score than in the placebo group (55 versus 49%, respectively). This effect lasted till Week 97. Interestingly, Hb level <9.0 g/dL at any point was significantly related with worse scores in all aspects of HRQOL, possibly due to intercurrent comorbidities.
The paper by de Goeij et al. published in this issue of the Journal give us interesting information about achieved Hb levels and quality of life of non-dialysis CKD patients. The two key strengths of this observational study are that data on quality of life were collected longitudinally and were analysed separately among old and young patients. The findings of randomized clinical trials suggest that patients having more comorbidities and treated with higher ESA doses are those more likely to be exposed to higher risk. In this perspective, the elderly population may be more likely to be affected by a higher burden of comorbidities and thus beneficiate less of anaemia correction for what concerns quality of life.
The results of this study clearly go in this direction, showing that in elderly patients who were prescribed ESA/iron therapy, high Hb levels (≥13 g/dL) were not associated with significant differences in physical and mental status compared with lower values (11–12 g/dL). Conversely, young patients prescribed ESA/iron therapy had a higher physical and mental summary score at high Hb levels. This finding is not unexpected, given that compared with younger patients, elderly ones often have a reduced level of activity. However, the effect of quality of life in different age groups has been little studied previously, despite its possible important implication in everyday clinical practice.
Another point confirmed by this study is that naturally occurring Hb levels and Hb target reached with ESA/iron treatment have a different meaning on patient outcome and quality of life. This seems particularly true for the elderly patients. Altogether, these findings stress the urge of treatment individualization in anaemic CKD patients, selecting different Hb targets in order to improve quality of life, without negatively affecting survival or comorbidities. Unfortunately, in this study the number of patients with very low Hb levels (<10 g/dL) was small (only 16 patients), making it impossible to clarify one of the major points of concern of the recent KDIGO guidelines referring to the treatment of anaemic CKD patients especially if not on dialysis, i.e. the long-term effects of keeping them at very low Hb levels without starting ESA therapy.
After 20 years from the introduction of ESAs in clinical practice, there is still a great uncertainty about the optimal Hb target to aim at in CKD patients. There is general agreement that partial anaemia correction improves outcome. The current evidence on hard end points indicates that intentionally aiming at Hb levels of 9.5–11.5 g/dL is associated with better outcomes compared with >13 g/dL. It is also clear that mild benefits on quality of life cannot justify a higher risk of cardiovascular events when exceeding intentionally the Hb value of 13 g/dL, particularly in elderly patients. The TREAT trial, despite its rigorous design, has left several uncertainties. Achieved Hb values in the control group were well over the rescue value of 9 g/dL, starting from 10.4 g/dL and finishing at 11.3 g/dL, with a mean value of 10.6 g/dL. This does not allow us to understand whether the effect of anaemia correction on quality of life may be more relevant when anaemia is symptomatic, i.e. when the patients are severely anaemic (Hb <10 g/dL); on the contrary, correcting mild anaemia may not cause a clinically significant effect because patients have already reached a plateau in their performance level. Another point to consider is that clinical trials report averages. The 'modest improvement in patient-reported fatigue' shown by the TREAT study may be more substantial in the single patient. Medical care needs scientific evidences, but it is wrong to assume that 'all individuals would weigh the risks and benefits and arrive at the same risk-benefit calculation'. In this respect, the patient's point of view is often very different from that of the doctor, since for patients, especially for those who have a disease with a relatively short life expectancy, quality of life may be of primary importance rather than its duration.
More reliable methods of assessing patient-related outcomes and functional status are now available. While the mean Hb level of the CKD population is constantly lowering worldwide, there is the urgent need of new, well-designed studies testing the effect of anaemia correction on more robust measures of quality of life. For the sake of evidence-based medicine, we are running the risk of forcing our patients to cope with a miserable quality of life despite the availability of an effective therapy.
Almost all of the papers dealing with chronic kidney disease (CKD) invariably start underlining that morbidity and mortality rate of this patient population is unacceptably high. Conversely, very few papers have considered quality of life and none selected quality of life as the primary end point. This reflects a clear bias for doctors and investigators, possibly because the treatment of CKD, especially in stage 5 on dialysis or with transplantation, is life-saving. Thus, the efforts of nephrologists were and are still concentrated in reducing the mortality and the morbidity of patients, without carefully considering quality of life. This attitude has been amplified by the influence of evidence-based medicine and regulatory authorities: the only end points that really matter are the hard-ones. Surrogate end points, especially when they are not easily quantifiable, are considered often insufficient to prove the efficacy of a given treatment. This process has markedly improved the scientific quality of clinical trials. However, we are reaching the paradox that no clinical trial is capable of proving efficacy on hard end points. Recently, almost all large, well-designed clinical trials testing a variety of treatments have given negative results. This has led physicians back to the importance of individualizing the cure of the single patient according to his characteristics and it is revaluating a comprehensive view of the possible associated benefits of a given treatment.
Apart from treating hypertension and dyslipidaemia, anaemia correction is an important potential area for reducing cardiovascular morbidity and mortality in CKD.
Before the availability of erythropoiesis stimulating agents (ESA), CKD patients, especially those on dialysis, were heavily anaemic and symptomatic. Blood transfusions partially and temporary alleviated symptoms, but at the price of the risk of viral infections, iron overload and sensitization, which could jeopardize the possibility of receiving a transplant.
Since the early 1990s, the introduction of ESA has revolutionized the quality of life of CKD patients, who previously had very limited possibility not only to maintain a job, but also to move: they could barely survive and did not enjoy life. Doctors very soon realized the potential of ESAs. These agents were very effective in increasing haemoglobin (Hb) levels even to normal values. A striking improvement in quality of life was the second most evident effect. However, despite this marked improvement in patient well-being, in 1989 the Food and Drug Administration (FDA) approved Epoetin alfa 'to elevate or maintain the red blood cell level and to decrease the need for transfusions', without even mentioning quality of life.
Observational data from large databases have invariably shown a clear relationship between anaemia and outcomes: the worse the anaemic status, the worse the patient mortality and morbidity. The possibility of improving survival and reducing comorbidities through anaemia correction became the aim of several clinical trials. Perhaps forgetting that mild anaemia may have a compensatory role in CKD patients, trials were designed to compare near-to-normal Hb target ranges with partial anaemia correction.
Back to the 1990s, the Normal Hematocrit study was stopped for futility, together with trend towards a greater risk of death or myocardial infarction in the patient group that was randomized to achieve a normal haematocrit.
The general feeling of nephrologists was that ESA treatment was possibly started too late, when the cardiovascular damage was too advanced to obtain benefit with anaemia treatment. However, subsequent trials, which enrolled non-dialysis, less compromised patients, gave disappointing results as well.
In 2009, the publication of the Trial to Reduce Cardiovascular Events with Aranesp® Therapy (TREAT) sanctioned the end of the dream of improving survival and the risk of cardiovascular events by means of complete anaemia correction. Despite neutral findings at the primary analysis, secondary analyses showed a higher risk of stroke and death for cancer in patients with a history of strokes and malignancies, respectively.
The wake of these findings, further strengthened by the publication of Kidney Disease: Improving Global Outcomes (KDIGO) guidelines (and possibly by changes in ESA reimbursement practice in the USA), has led to a progressive reduction of ESA doses and achieved Hb values. Today, an increasing percentage of CKD patients experience severe anaemia and suffer those anaemia-related symptoms that were only observed in the pre-ESA era and always more frequently receive blood transfusions.
In this uncertain scenario, quality of life has to be reconsidered and weighted with the risk of adverse events in the single patient.
According to the common experience of physicians in everyday clinical practice, early trials using erythropoietin showed improvements in multiple domains and measures of quality of life. In particular, data of three of these early trials were put together to assess the effect of erythropoietin treatment on quality of life in over 1500 patients. Haemodialysis patients treated with rHuEPO had a significant improvement in objective (functional ability and health status) and subjective (well-being, life satisfaction, psychological aspects and happiness) quality of life indicators at 10 months. A newer analysis of the Canadian Erythropoietin Study also showed a significant increase in exercise tolerance and patient-reported physical function in the epoetin-alfa treated patients compared with those receiving placebo. Improvements were seen at 2 months and were maintained through Months 4 and 6.
The interpretation of these studies was complicated by methodological limitations (no control group) and by the fact that instruments for measuring quality of life and achieved Hb levels were heterogeneous. The median follow-up was quite short, not giving information whether changes in quality of life are long lasting. Indeed, these data were reviewed by the FDA and considered not convincing.
Several meta-analyses tried to summarize the effect of anaemia correction on quality of life in both dialysis and non-dialysis patients and showed a consistent and positive impact on several aspects of quality of life. However, they inevitably suffered the methodological limitations of the included studies.
Recent clinical trials used more standardized measures of quality of life. The Cardiovascular Risk Reduction by Early Anaemia Treatment (CREATE) study, which was a large, open-label study, only tested quality of life with the Short-Form Health Survey (SF-36) questionnaire. At 1 year, anaemia correction towards Hb levels >13 g/dL was associated with a moderate effect on quality of life compared with partial correction (Hb 10.5–11.5 g/dL) with regard to general and mental health, physical function, physical role, social function and vitality. At the end of the second year, the significant difference between groups was maintained only for general health and vitality. The Correction of Haemoglobin and Outcomes in Renal Insufficiency (CHOIR) study used a larger set of instruments to test quality of life [Linear Analogue Self-Assessment (LASA), the Kidney Disease Questionnaire (KDQ) and SF-36]. Both patients in the low- and high Hb group reported an improvement in energy and physical function score without any difference between the two groups. A higher burden of comorbidities and a small difference in the mean achieved Hb levels between the two groups (1.3 g/dL) may possibly explain why these findings are not concordant with those of the CREATE trial.
The TREAT is the largest double-blind, placebo controlled trial in the field. In this study, QOL was measured with Functional Assessment of Cancer Therapy (FACT)-Fatigue, Short Form-36, and EuroQol scores over a follow-up period of nearly 2 years. Overall, health-related quality of life (HRQOL) improved more in the darbepoetin compared with the placebo group. However, at both the adjusted and unadjusted analysis, only the FACT-fatigue score was significantly improved in the patients treated with darbepoetin compared with placebo. This effect was judged of modest clinical significance. Of note, a higher percentage of patients in this group had a ≥3-point increase (i.e. a clinically meaningful change) of the score than in the placebo group (55 versus 49%, respectively). This effect lasted till Week 97. Interestingly, Hb level <9.0 g/dL at any point was significantly related with worse scores in all aspects of HRQOL, possibly due to intercurrent comorbidities.
The paper by de Goeij et al. published in this issue of the Journal give us interesting information about achieved Hb levels and quality of life of non-dialysis CKD patients. The two key strengths of this observational study are that data on quality of life were collected longitudinally and were analysed separately among old and young patients. The findings of randomized clinical trials suggest that patients having more comorbidities and treated with higher ESA doses are those more likely to be exposed to higher risk. In this perspective, the elderly population may be more likely to be affected by a higher burden of comorbidities and thus beneficiate less of anaemia correction for what concerns quality of life.
The results of this study clearly go in this direction, showing that in elderly patients who were prescribed ESA/iron therapy, high Hb levels (≥13 g/dL) were not associated with significant differences in physical and mental status compared with lower values (11–12 g/dL). Conversely, young patients prescribed ESA/iron therapy had a higher physical and mental summary score at high Hb levels. This finding is not unexpected, given that compared with younger patients, elderly ones often have a reduced level of activity. However, the effect of quality of life in different age groups has been little studied previously, despite its possible important implication in everyday clinical practice.
Another point confirmed by this study is that naturally occurring Hb levels and Hb target reached with ESA/iron treatment have a different meaning on patient outcome and quality of life. This seems particularly true for the elderly patients. Altogether, these findings stress the urge of treatment individualization in anaemic CKD patients, selecting different Hb targets in order to improve quality of life, without negatively affecting survival or comorbidities. Unfortunately, in this study the number of patients with very low Hb levels (<10 g/dL) was small (only 16 patients), making it impossible to clarify one of the major points of concern of the recent KDIGO guidelines referring to the treatment of anaemic CKD patients especially if not on dialysis, i.e. the long-term effects of keeping them at very low Hb levels without starting ESA therapy.
After 20 years from the introduction of ESAs in clinical practice, there is still a great uncertainty about the optimal Hb target to aim at in CKD patients. There is general agreement that partial anaemia correction improves outcome. The current evidence on hard end points indicates that intentionally aiming at Hb levels of 9.5–11.5 g/dL is associated with better outcomes compared with >13 g/dL. It is also clear that mild benefits on quality of life cannot justify a higher risk of cardiovascular events when exceeding intentionally the Hb value of 13 g/dL, particularly in elderly patients. The TREAT trial, despite its rigorous design, has left several uncertainties. Achieved Hb values in the control group were well over the rescue value of 9 g/dL, starting from 10.4 g/dL and finishing at 11.3 g/dL, with a mean value of 10.6 g/dL. This does not allow us to understand whether the effect of anaemia correction on quality of life may be more relevant when anaemia is symptomatic, i.e. when the patients are severely anaemic (Hb <10 g/dL); on the contrary, correcting mild anaemia may not cause a clinically significant effect because patients have already reached a plateau in their performance level. Another point to consider is that clinical trials report averages. The 'modest improvement in patient-reported fatigue' shown by the TREAT study may be more substantial in the single patient. Medical care needs scientific evidences, but it is wrong to assume that 'all individuals would weigh the risks and benefits and arrive at the same risk-benefit calculation'. In this respect, the patient's point of view is often very different from that of the doctor, since for patients, especially for those who have a disease with a relatively short life expectancy, quality of life may be of primary importance rather than its duration.
More reliable methods of assessing patient-related outcomes and functional status are now available. While the mean Hb level of the CKD population is constantly lowering worldwide, there is the urgent need of new, well-designed studies testing the effect of anaemia correction on more robust measures of quality of life. For the sake of evidence-based medicine, we are running the risk of forcing our patients to cope with a miserable quality of life despite the availability of an effective therapy.
Source...