Description of the condition
Erythropoiesis-stimulating agents (ESAs) are perhaps the most rigorously tested group of drugs in nephrology. Since the introduction of ESAs, there have been substantial reductions in blood transfusion requirements among patients living with chronic kidney disease (CKD) (Eschbach 1989).
A systematic review of 14 randomised controlled and uncontrolled trials in pre-dialysis CKD patients demonstrated that treatment of anaemia with ESAs improved energy levels and physical function (Gandra 2010). Unfortunately, a considerable proportion of these patients exhibited suboptimal haematologic response to ESA (Benz 1999; Valderrabano 1996).
There are several known causes of suboptimal response to ESA. These include deficiencies in iron, vitamin B
ESA treatment used to target high haemoglobin levels in people with CKD is associated with deleterious (Phrommintikul 2007) or neutral (Palmer 2010) impacts on survival and increased risks of stroke, vascular access thrombosis and hypertension without any reduction in cardiovascular events (Palmer 2010; Phrommintikul 2007).
Although RCTs and systematic reviews consistently show more harm than benefit associated with higher haemoglobin targets for ESA treatment (Besarab 1998; Palmer 2010; Pfeffer 2009; Phrommintikul 2007; Singh 2006), secondary analyses of RCTs and observational studies have demonstrated that poor response to ESA treatment rather than achieved high haemoglobin, may be responsible for the observed suboptimal outcomes in people with CKD (Kilpatrick 2008; Messana 2009; Regidor 2006; Solomon 2010; Szczech 2008).
These studies also showed that patients who required higher doses of ESA experienced increased mortality at any haemoglobin level, and that patients who achieve target haemoglobin levels had better outcomes than those who did not (Badve 2011). Therefore, therapies targeting ESA resistance could be a promising treatment strategy in CKD anaemia management.
Description of the intervention
Although there is no effective treatment for patients with ESA-resistant anaemia at present, a number of interventions such as L-carnitine, ascorbic acid, oxpentifylline, androgens and statins have been investigated.
This review looked at the benefits and harms of any intervention used in the treatment of ESA-resistant anaemia in people with end-stage kidney disease (ESKD) who were receiving dialysis.
Criteria for considering studies for this review
Types of studies
All randomised controlled trials (RCTs) and quasi-RCTs (studies in which allocation to treatment was obtained by alternation, use of alternate medical records, date of birth or other predictable methods) looking at interventions for the treatment of ESA-resistant anaemia in people with ESKD were included in our review.
Types of participants
- Adults and children with ESKD (chronic kidney disease (CKD) stage 5 or pre-dialysis) or those receiving dialysis (either haemodialysis or peritoneal dialysis).
- Adults and children with ESKD receiving any type of ESA for anaemia (anaemia defined as haemoglobin < 110 g/L or as defined by the investigators).
- Evidence of ESA resistance, defined as failure to achieve or maintain target range haemoglobin/haematocrit levels in spite of appropriate ESA doses (erythropoietin ≥ 450 U/kg/wk intravenous administration or ≥ 300 U/kg/wk for subcutaneous administration or darbepoetin ≥ 1.5 µg/kg/wk) (KDOQI 2001; Locatelli 2004). This inclusion criterion was amended after publication of the protocol of this systematic review because only one eligible study was found. Extended inclusion criteria were studies that defined ESA-hyporesponsive state as failure to achieve or maintain target haemoglobin/haematocrit in spite of the following doses of the ESA: erythropoietin dosage ≥ 300 and ≥150 U/kg/wk for IV administration; or ≥ 200 and ≥100 U/kg/wk for subcutaneous administration; or darbepoetin dosage ≥ 1.0 µg/kg/wk).
- All known causes of ESA-resistance (such as iron deficiency, vitamin B
12deficiency, folate deficiency, infection, chronic inflammatory state, neoplasia, severe hyperparathyroidism, aluminium intoxication, inadequate dialysis, myelosuppressive agents, haemoglobinopathies, myelodysplasia and antibody-mediated pure red cell aplasia) must have been ruled out.
- Studies performed in kidney transplant recipients were excluded.
Types of interventions
Any potential intervention used to treat ESA-resistance, such as L-carnitine, ascorbic acid, oxpentifylline, androgens, and statins, were included in this review.
Types of outcome measures
- All-cause mortality
- Cardiovascular mortality
- Non-fatal cardiovascular events
- Number of patients achieving target haemoglobin/haematocrit
- Difference or changes in haemoglobin or haematocrit between intervention and control groups at study end
- Difference or changes in ESA dose between intervention and control groups at study end
- Blood transfusion requirements
- Quality of life
- Any reported adverse events
- Differences or changes in inflammatory biomarkers between intervention and control groups at study end
- Differences or changes in biomarkers of oxidative stress between intervention and control groups at study end.
Search methods for identification of studies
We searched the Cochrane Renal Group's specialised register 18th March 2013 through contact with the Trials' Search Co-ordinator using search terms relevant to this review.
The Cochrane Renal Group’s Specialised Register contains studies identified from:
- Monthly searches of the Cochrane Central Register of Controlled Trials (CENTRAL)
- Weekly searches of MEDLINE OVID SP
- Handsearching of renal-related journals and the proceedings of major renal conferences
- Searching of the current year of EMBASE OVID SP
- Weekly current awareness alerts for selected renal journals
- Searches of the International Clinical Trials Register (ICTRP) Search Portal and ClinicalTrials.gov.
Studies contained in the specialised register are identified through search strategies for CENTRAL, MEDLINE and EMBASE based on the scope of the Cochrane Renal Group. Details of these strategies, as well as a list of handsearched journals, conference proceedings and current awareness alerts, are available in the specialised register section of information about the Cochrane Renal Group.
See Appendix 1 for search terms used in strategies for this review.
Searching other resources
- Reference lists of clinical practice guidelines, review articles and relevant studies.
- Relevant missing or incomplete or unpublished data from the clinical studies were requested from the respective investigators/ authors by written correspondence.
Data collection and analysis
Selection of studies
The search strategy described was used to obtain titles and abstracts of studies relevant to the review. Titles and abstracts were screened independently by three authors, who discarded studies that were not applicable. However, studies and reviews that potentially included relevant data or study information were retained initially. The same three authors independently assessed retrieved abstracts, and if necessary the full text, of these studies to determine which studies satisfied the inclusion criteria.
Data extraction and management
Data extraction was carried out independently by two authors using standard data extraction forms. Studies reported in non-English language journals was to be translated before assessment. Where more than one publication of one study existed, reports were grouped together and the publication with the most complete data was used in the analyses. Where relevant outcomes were only published in earlier versions, these data were used. Any discrepancies between published versions was to be highlighted. Disagreements were resolved by consensus.
Assessment of risk of bias in included studies
- Was there adequate sequence generation (selection bias)?
- Was allocation adequately concealed (selection bias)?
- Was knowledge of the allocated interventions adequately prevented during the study (detection bias)?
- Participants and personnel
- Outcome assessors
- Were incomplete outcome data adequately addressed (attrition bias)?
- Are reports of the study free of suggestion of selective outcome reporting (reporting bias)?
- Was the study apparently free of other problems that could put it at a risk of bias?
Measures of treatment effect
For dichotomous outcomes (all-cause mortality, cardiovascular mortality, non-fatal cardiovascular events, number of patients achieving haemoglobin/haematocrit targets, number of patients requiring hospitalisation, number of patients requiring blood transfusions, number of patients with medication-related adverse effects), results were expressed as risk ratios (RR) with 95% confidence intervals (CI). For continuous data (haemoglobin, haematocrit, iron studies, ESA dosage, iron dosage, hospitalisation days, quality of life scores, inflammatory biomarkers, biomarkers of oxidative stress), results were expressed as mean difference (MD).
Dealing with missing data
We planned that any further information required from the original author was to be requested by written correspondence, and any relevant information obtained was be included in the review. Evaluation of important numerical data such as screened, randomised patients as well as intention-to-treat (ITT), as-treated and per-protocol (PP) population was performed.
Assessment of heterogeneity
Heterogeneity was to be analysed using a Chi² test on N-1 degrees of freedom, with an alpha of 0.05 used for statistical significance and with the I² test (Higgins 2003). I² values of 25%, 50% and 75% correspond to low, medium and high levels of heterogeneity.
Data were to be pooled using the random-effects model.
Description of studies
Results of the search
We identified 533 abstracts using the search strategy described (Figure 1). After screening titles and abstracts, 99 reports were selected for full text review. Only two studies (Attallah 2006; Ayli 2004) met our inclusion criteria, and of these, one investigated our extended inclusion criterion of ESA hyporesponsive state (Ayli 2004).
|Figure 1. The PRISMA flow chart showing selection of studies|
We considered inclusion of a study that applied our extended inclusion criterion of ESA-hyporesponsive state (Sezer 2002). In this study, participants in both arms received the investigational drug (vitamin C) in the first study phase (eight weeks). Non-responders were excluded at the end of the first phase. During the second phase, remaining participants were randomised to receive either the investigational drug at a reduced frequency or no study drug for another eight weeks. Since the investigators did not define 'non-responder', and there was a strong possibility of carry over effect of vitamin C administered before randomisation, the study was excluded from this systematic review.
Two studies met our inclusion criteria.
- Attallah 2006 enrolled 42 haemodialysis patients and compared IV vitamin C given at each dialysis session to no treatment.
- Ayli 2004 enrolled 48 haemodialysis patients and compared high-flux versus low-flux dialysis membranes
We excluded 68 studies after full-text review: six were not randomised; 58 included participants who did not have ESA resistance; two included iron deficient participants who lacked true ESA resistance; and two studies did not use ESA in the control arm.
Risk of bias in included studies
Incomplete outcome data
All participants were followed for the entire study period and accounted for in both studies. Attrition bias arising from incomplete outcome reporting was deemed to be low risk.
Neither study reported proportions of participants in each study arm who achieved haemoglobin target levels. The risk of reporting bias in both was therefore unclear.
Other potential sources of bias
Both studies were judged to be at high risk of other potential sources of bias due to single-centre study design and exclusion of patients on peritoneal dialysis.
Effects of interventions
All-cause and cardiovascular mortality
No deaths were reported in either study.
Non-fatal cardiovascular events
Ayli 2004 did not report non-fatal cardiovascular events.
Participants achieving target haemoglobin or haematocrit
Neither study reported the proportions of participants who achieved target haemoglobin or haematocrit levels.
Requirement of blood transfusions
Attallah 2006 reported no participants included in the final analysis required blood transfusion. However, one participant from the control group was excluded from the final analysis because of the need for a blood transfusion due to a significant upper gastrointestinal bleed.
Ayli 2004 did not report need for blood transfusion.
Ayli 2004 did not report hospitalisations.
Medication-related adverse events
Haematology and biochemistry results
Both studies reported significantly higher haemoglobin levels in the treatment groups compared to the control groups ( Analysis 2.1.1: MD 0.9 g/dL, 95% CI 0.38 to 1.42; Attallah 2006); ( Analysis 2.1.2: MD 1.9 g/dL, 95% CI 1.64 to 2.16; Ayli 2004).
Attallah 2006 did not report data on participants' haematocrit levels. Ayli 2004 reported that among interventional arm participants haematocrit was significantly higher than those in the control arm ( Analysis 2.2: MD 6.8%, 95% CI 5.67 to 7.93).
Transferin saturation (TSAT)
Attallah 2006 reported that TSAT was significantly higher in interventional than control arm participants ( Analysis 2.3.1: MD 8.00%, 95% CI 6.22 to 9.78). There was no significant difference in TSAT between study arms reported by Ayli 2004 ( Analysis 2.3.2: MD 1.30%, 95% CI -3.99 to 6.59).
Attallah 2006 reported that ferritin was significantly higher among interventional than control arm participants ( Analysis 2.4.1: MD 8.00 ng/mL, 95% CI -85.51 to 101.51). There was no significant difference between study arms reported by Ayli 2004 ( Analysis 2.4.2: MD -3.00 ng/mL, 95% CI -43.46 to 37.46).
Haemoglobin content in reticulocytes (CHr)
Inflammatory biomarkers: C-reactive protein
Attallah 2006 reported C-reactive protein was significantly lower in vitamin C group compared to the control group ( Analysis 2.6.1: MD -1.20 mg/dL, 95% CI -1.69 to -0.71). There was no significant difference between study arms in C-reactive protein reported by Ayli 2004 ( Analysis 2.6.2: MD -0.4 mg/dL, 95% CI -3.0 to 2.2).
Markers of oxidative stress
ESA and intravenous iron doses
Intravenous iron therapy dose
Attallah 2006 reported that there was no significant difference in intravenous iron therapy dose between the study arms ( Analysis 3.2: MD -0.20 mg/wk, 95% CI -16.15 to 15.75). Ayli 2004 did not report on intravenous iron therapy dose.
Quality of life scores
The results of this systematic review highlight the absence of adequately powered randomised controlled trials (RCT) examining the effect of various interventions to treat ESA hyporesponsiveness. We found that there was insufficient and inadequate evidence to recommend any intervention to ameliorate ESA-hyporesponsiveness.
We identified only one RCT that defined ESA-hyporesponsiveness as intravenous EPO dose ≥ 450 U/kg/wk (Attallah 2006). When inclusion criteria were extended to include subcutaneous EPO dose ≥ 200 U/kg/wk, another study, Ayli 2004, was found to be eligible for inclusion.
In relation to intravenous vitamin C therapy, Attallah 2006 demonstrated increases in haemoglobin, haemoglobin content in reticulocytes, and transferin saturation; and reductions in erythropoietin dose and C-reactive protein. Ayli 2004 reported that use of high-flux dialyser for six months was associated with improvement in haemoglobin, but there was no effect on C-reactive protein or iron studies. Both Attallah 2006 and Ayli 2004 were single-centre studies and included 42 and 48 participants respectively. The studies included only haemodialysis patients, and hence, results may not be generalisable to CKD patients not yet on dialysis, those on peritoneal dialysis, or in settings where patient populations differ.
There is no single widely accepted definition of ESA resistance. KDOQI has defined ESA resistance as failure to achieve haemoglobin 11 g/dL with ESA dose equivalent to epoetin greater than 500 IU/kg/wk (KDOQI 2006). Publication of KDIGO anaemia guidelines is expected this year. As yet, there have been no RCTs performed explicitly in patients with ESA resistance as defined by KDOQI.
In the Normal Haematocrit Cardiac Trial, more participants in the normal haematocrit group reached the primary endpoint (composite of death and non-fatal myocardial infarction) with mean erythropoietin doses of 440 IU/kg/wk, which is lower than the KDOQI definition (Besarab 1998). In the CHOIR trial, it was reported that ESA dose > 20,000 IU/wk was associated with increased risk of death, congestive heart failure, stroke, and myocardial infarction (Szczech 2008).
Several observational studies have suggested a linear association between ESA dose and adverse outcomes (Brookhart 2010; Messana 2009; Regidor 2006; Zhang 2004; Zhang 2009). There is substantial variability in the reporting of ESA dose, such as IU/kg/wk, IU/wk, or ESA dose normalised to haemoglobin level. Therefore, the current KDOQI definition of ESA resistance needs to be revised, and the new definition should be based on ESA-resistance index (ERI) rather than ESA dose to bring uniformity in reporting.
The revised inclusion criteria of the ongoing HERO Study are ESA-resistance index ≥ 1.0 IU/kg/wk/haemoglobin for epoetin-treated patients and ≥ 0.005 µg/kg/wk/g haemoglobin for darbepoetin-treated patients (Johnson 2008). Table 1 presents current definitions of ESA resistance.
An emerging body of evidence indicates more harm than benefit from targeting higher haemoglobin levels with ESA therapy. Patients who needed higher doses of ESA experienced increased mortality at any haemoglobin level, and patients who achieved target haemoglobin levels had better outcomes than those who did not.
Further RCTs are needed urgently to consider the clinical impacts of therapies purported to reduce ESA resistance.
Implications for practice
Based on two small, single-centre studies, there was inadequate evidence to recommend any intervention to ameliorate ESA-hyporesponsiveness.
Implications for research
Adequately powered multicentre RCTs involving a wide range of CKD patients receiving ESA therapy should be conducted as a priority. In addition to those on haemodialysis, future RCTs should include pre-dialysis CKD patients as well people receiving peritoneal dialysis.
Future studies should focus on true ESA responsiveness rather than a haemoglobin-targeted approach. Importantly, these studies should also include cost-effectiveness and economic analyses.
The authors would like to acknowledge Narelle Willis and Ruth Mitchell from the Cochrane Renal Group for their assistance. The authors would also like to thank the referees for their editorial advice during the preparation of this review.
Data and analyses
- Top of page
- Authors' conclusions
- Data and analyses
- Contributions of authors
- Declarations of interest
- Sources of support
- Differences between protocol and review
- Index terms
Appendix 1. Electronic search strategies
Appendix 2. Risk of bias assessment tool
Contributions of authors
- Write the protocol: SB, DF, EB, CH, DJ, IM, AC, VP
- Study selection: SB, CH, DJ
- Extract data from studies: SB, DJ
- Enter data into RevMan: SB, DJ
- Data analysis: SB, DF, EB
- Interpret the analysis: SB, DJ
- Draft the final review: SB, DJ
- Disagreement resolution: DF, EB, CH, IM, AC, VP
- Update the review: SB, DJ
Declarations of interest
- Dr Sunil V Badve, Elaine Beller and Daniel P Francis have no conflicts of interest to declare.
- Associate Professor Carmel Hawley has received consulting fees from Amgen and Janssen-Cilag; research grants from Amgen, Roche and Janssen-Cilag; and speakers’ honoraria from Amgen.
- Professor Alan Cass is the recipient of a NHMRC Senior Research Fellowship. He has received speaker's honoraria and research grants from Janssen-Cilag, Amgen and Roche.
- Associate Professor Vlado Perkovic has received speakers’ honoraria from Roche and research grants from Johnson and Johnson Pharmaceutical Research & Development and Roche.
- Professor Iain C. Macdougall has received consultant fees, research grants, and/or lecture fees from Amgen, Ortho biotech, Roche, Affymax, Takeda, Hospira, and Sandoz.
- Professor David Johnson has received speakers' honoraria, consultancy fees and research grants from Janssen-Cilag, Amgen and Roche. He has received fees for organising education from Amgen and Janssen-Cilag. He has received consultancy fees from Pfizer. He is also the Principal Investigator in the HERO Trial, a randomised, double-blind, placebo-controlled trial of oxpentifylline in the treatment of erythropoietin stimulating agent hyporesponsiveness. Professor Alan Cass and Associate Professor Carmel Hawley are the members of the Trial Management Committee of the HERO trial.
Sources of support
- Australasian Kidney Trials Network, School of Medicine, University of Queensland, Australia.
- Princess Alexandra Hospital, Woolloongabba, QLD, Australia.
- No sources of support supplied
Differences between protocol and review
In the protocol for this review, we had planned that one of our inclusion criteria would define ESA resistance. Evidence of ESA-resistance, defined as failure to achieve or maintain target range haemoglobin/haematocrit levels in spite of appropriate doses of the ESA (erythropoietin dose ≥ 450 U/kg/wk intravenous administration or ≥ 300 U/kg/wk for subcutaneous administration or darbepoetin dosage ≥ 1.5 µg/kg/wk) (KDOQI 2001; Locatelli 2004) was to be applied. This inclusion criterion was amended because only one eligible study was found.
Medical Subject Headings (MeSH)
*Renal Dialysis; Anemia [blood; *drug therapy]; Drug Resistance; Erythropoiesis [*drug effects]; Erythropoietin [*administration & dosage]; Hematocrit; Kidney Failure, Chronic [*complications; therapy]; Randomized Controlled Trials as Topic
MeSH check words
* Indicates the major publication for the study