Maintenance of target hemoglobin level in stable hemodialysis patients constitutes a theoretical task: a historical prospective study

Transcription

1 original article & 2008 International Society of Nephrology Maintenance of target hemoglobin level in stable hemodialysis patients constitutes a theoretical task: a historical prospective study José M. Portolés 1,10,11,Ángel L.M. de Francisco 2,10,11, José L. Górriz 3,11, Alberto Martínez-Castelao 4,11, Juan M. López-Gómez 5,11, Manuel Arias 2,10,11, Juan J. de la Cruz 6, Aleix Cases 7,10,11, Evaristo Fernández 8,11 and Pedro Aljama 9,10,11 1 Service of Nephrology, Fundación Hospital Alcorcón, Madrid, Spain; 2 Service of Nephrology, Hospital Universitario Valdecilla, Santander, Spain; 3 Service of Nephrology Hospital Universitario Dr Peset, Valencia, Spain; 4 Service of Nephrology, Hospital Universitario Bellvitge, L Hospitalet de Llobregat, Barcelona, Spain; 5 Service of Nephrology, Hospital Universitario Gregorio Marañón, Madrid, Spain; 6 Department of Preventive Medicine and Public Health, Autonomous University of Madrid, Madrid, Spain; 7 Nephrology Unit, Hospital Clínic, Universitat de Barcelona, Barcelona, Spain; 8 Service of Nephrology, Hospital Puerto Real, Cádiz, Spain; 9 Service of Nephrology, Hospital Universitario Reina Sofía, Córdoba, Spain and 10 REDinREN (Instituto de Salud Carlos III, Red 6/0016), Madrid, Spain Maintenance of target hemoglobin (Hb) values in hemodialysis patients treated with erythropoiesis-stimulating agents (ESAs) remains difficult. We examined Hb variability in the clinical setting in hemodialysis patients. Hemodialysis patients treated with ESAs who maintained the recommended Hb range of g per 100 ml over 3 months and were not admitted to hospital, did not require transfusion, and did not experience any major clinical event during this period were followed prospectively for 1 year. Anemia events, Hb variation events (any value out of ±1.5 g per 100 ml of the median Hb level in the total follow-up period for the individual patient), risk factors for anemia, and Hb variation events were assessed. We studied 420 patients (63% males, mean age 61 years), 222 received short-acting erythropoietin (EPO) and 198 long-acting darbepoetin. A total of 4654 blood samples (mean 11.1 per patient-year) were analyzed. Only 3.8% of patients were maintained within the target Hb levels (11 13 g per 100 ml) during 1 year. Hb variation events occurred in 20.8% of laboratory values and anemia events in 14.7%, with a median time to the first event of 3 months. Treatment with short-acting EPO (vs longacting darbepoetin), change of ESA dose in the previous visit, resistance index, and hospitalization were significant risk factors for both anemia events and Hb variation events. Our results show that Hb values are rarely maintained within the recommended guidelines even in more stable hemodialysis patients. Hb variability is frequently associated with clinical events or ESA dose changes. Long-acting darbepoetin achieved better Hb stability than short-acting EPO. ; doi: /ki Correspondence: José M. Portolés, Service of Nephrology, Fundación Hospital Alcorcón, Avda. Villaviciosa 1, E Alcorcón, Madrid, Spain. jmportoles@fhalcorcon.es 11 Members of the Anemia Study Group Spanish Society of Nephrology. KEYWORDS: erythropoiesis-stimulating agents; darbepoetin; hemodialysis; hemoglobin variability Anemia is common in patients on hemodialysis. Hemoglobin (Hb) levels o11 g per 100 ml are associated with an impairment in quality of life, reduced energy, decreased exercise capacity, increased costs, morbidity, and mortality. 1 3 The cause of anemia in these patients is mainly related to a deficiency in the synthesis of endogenous erythropoietin (EPO). 4 Although treatment with erythropoiesis-stimulating agents (ESAs) is effective to reverse anemia, considerable controversy exists with regard to the optimal Hb level and the upper limit of the desired range. 5 In fact, after the publication of two major studies of recombinant human EPO in chronic kidney disease (CKD), showing that early complete correction of anemia did not reduce the risk of cardiovascular events, 6,7 anemia guidelines on Hb target have recommended a general range of g per 100 ml and never over 13 g per 100 ml in dialysis and non-dialysis patients with CKD receiving ESA therapy. 8 However, the need to maintain Hb levels within a narrow target range may result in an insufficient ESA treatment, with increasing time for patients on subtargeted Hb values. 9 Earlier studies usually carried out in the United States have shown that a relatively small percentage of ESA-treated CKD patients maintain Hb levels within the guidelinerecommended goals during a reasonable period of time. Fluctuations in Hb levels are very common, and different studies have explored Hb variability and the phenomenon of Hb cycling in individual patients. 10,11 There is an increasing interest not only in anemia correction but also to maintain Hb levels within a stability range gradually narrower. However, standard practices in anemia management differ largely among countries. 12 Moreover, clinicians have S82

2 JM Portolés et al.: Maintaining hemoglobin target in hemodialysis patients o r i g i n a l a r t i c l e difficulties in maintaining Hb concentrations within the target and frequently have doubts about the influence of a specific target range on the risk of adverse outcomes for a given patient. Factors associated with successful anemia management and Hb stability have been poorly defined. 13,14 Therefore, a 1-year prospective study was conducted to assess Hb variability under daily practice conditions in hemodialysis patients treated with ESA. RESULTS During the study period, 420 consecutive hemodialysis patients (63.1% men, mean age 61.0±14.7 years, mean weight 64.8±11.5 kg) were included in the prospective cohort. Almost half of the patients (49.5%) were treated at hospital hemodialysis units and the other half at dialysis centers. Main causes of CKD were glomerular disease (17.9%), diabetes mellitus (16.7%), ischemic nephropathy (16.2%), interstitial nephropathy (14.5%), polycystic kidney disease (8.7%), and unknown in 18.1% of the patients. At the time of inclusion, 12.5% of the patients were receiving dialysis by catheter, 75.0% with native arteriovenous fistula and 12.6% with a synthetic graft (PTFE Gore-tex). A total of 4654 blood samples (mean per patientyear) were analyzed. Mean Hb value was 12.2±1.3 g per 100 ml, mean ferritin level 470±301.7 ng/ml, mean serum iron 104±11.8 mg per 100 ml, mean transferrin saturation index 28.4±12.7%, and mean C-reactive protein 7.6±16.9 pg per 100 ml. The mean resistance to ESA index (ERI) was 12.3±4.8 IU/week/kg/g per 100 ml Hb. At the start of the study, 52.8% of the patients were receiving treatment with EPO at a mean dose of 9698±7388 U/week by the intravenous route and the remaining 47.1% long-acting darbepoetin at a mean dose of 26.7±38.0 mg/week (94.4% intravenous). All patients received iron supplements with a mean dose of 416.2±447.4 mg/month intravenously. As shown in Table 1, patients treated with EPO compared with those given long-acting darbepoetin were significantly older, less frequently dialyzed at the hospital, and received dialysis more frequently by synthetic graft and less frequently by catheter. The mean number of ESA dose adjustment during the follow-up period per patient was 2.6±2.3 (dose increase in 50.1%), with a mean change regarding the previous dose of 37.4±16.8%. In all, 13.8% of the patients required one dose change, 15% two dose changes, 12% three dose changes, and 34.3% of patients four or more dose changes. No significant differences between patients treated with EPO and those treated with darbepoetin were observed. A total of eight transfusions were administered and 130 hospital admissions were recorded during the follow-up (hospital admission rate, 0.34; 95% confidence interval (CI); per patient-year). Achievement of Hb target goals Compliant patients accounted for 3.8% of cases (95% CI, ), compliant patients with Hb increases 32.9% (95% CI, ), compliant patients with Hb decreases 10.5% (95% CI, ), and patients with Hb instability 52.9% (95% CI, ). The characteristics of these four groups of patients are shown in Table 2. Patients who maintained Hb on target (11 13 g per 100 ml) during 1 year were younger, with a higher percentage of native arteriovenous fistula and lower prevalence of diabetes mellitus as the cause of CKD than the remaining patients. Patients with Hb instability showed a significantly higher mean number of ESA dose change per month than the remaining groups. Hb variation events and anemia events The target range of g per 100 ml was documented in 57.5% of Hb measurements for a mean of almost 7 months. A decrease of greater than 1.5 g per 100 ml in the Hb level compared with the previous visit was recorded in 8.4% of the monthly visits, and an increase of greater than 1.5 g per 100 ml in 6.9%. On the other hand, Hb levels o11 g per 100 ml were recorded in 14.7% of measurements. Hemoglobin variation events (any value out of ±1.5 g per 100 ml of the median Hb level for the individual patient) Table 1 Characteristics of patients according to ESA treatment ESA treatment Variable Short-acting rhuepo (EPO) Long-acting darbepoetin P-value Total patients Age, years, mean (s.d.) 62.9 (14.0) 58.9 (15.1) Dialysis at the hospital (%) o0.001 Diabetes as cause of CKD, no. (%) 48 (21.6) 32 (16.2) Initial vascular access (%) Catheter Synthetic graft (PTFE) ESA dose change/year, mean (s.d.) 2.5 (2.2) 2.7 (2.3) ESA dose, U/week, mean (s.d.) a 9698 (7388) 8681 (4680) CKD, chronic kidney disease; EPO, erythropoietin; ESA, erythropoiesis-stimulating agent; rhuepo, recombinant human EPO. a 1 mg of darbepoetin-a=200 IU of epoetin-a or -b. S83

3 o r i g i n a l a r t i c l e JM Portolés et al.: Maintaining hemoglobin target in hemodialysis patients Table 2 Characteristics of patients according to compliance with Hb target goals during the 1-year follow-up period Variable Compliant patients a Compliant patients with Hb increases b Compliant patients with Hb decreases c Patients with Hb instability d Total patients Age, years, mean (s.d.) 58.2 (16.2) 62.2 (13.8) 63.2 (13.8) 60.4 (15.8) Type of center, no. (%) Hospital 8 (50) 70 (50.7) 26 (59.1) 104 (46.8) Dialysis center 8 (50) 68 (49.3) 18 (40.9) 118 (53.2) Initial vascular access AVF 14 (87.5) e 98 (71.5) 32 (72.7) 171 (77.4) Catheter 1 (6.3) 16 (11.7) 5 (11.4) 30 (77.4) ESA dose, U/week, mean (s.d.) f 8702 (3730) 8116 (4777) 8809 (7807) 9099 (6299) ESA dose change/month, mean (s.d.) 2.8 (2.5) 2.1 (2.1) 1.5 (1.7) 3.1 (2.3) o0.001 AVF, arteriovenous fistula; ESA, erythropoiesis-stimulating agent; Hb, hemoglobin. a All values within g per 100 ml. b All values 411 g per 100 ml and some 413 g per 100 ml. c All values o13 g per 100 ml and some o11 g per 100 ml. d Some values o11 and 413 g per 100 ml. e Compliant patients vs the remaining groups, Po0.05. f 1 mg of darbepoetin-a=200 IU of epoetin-a or -b. P-value Table 3 Factors associated with the appearance of Hb variation events and anemia events at each monthly visit Hb variation events a Anemia events b Variable Present Absent P-value Present Absent P-value Long-acting darbepoetin vs EPO (%) o o0.001 Change of ESA dose vs maintenance of the same dose (%) o o0.001 ESA dose (1000 U/week) 9.79 (7.44) 8.53(6.34) o (7.86) 8.44 (6.29) o0.001 ERI, U/week/g Hb, mean (s.d.) 15.1 (5.2) 11.6 (4.8) o (6.8) 11.1(4.9) o0.001 Hospital admission, mean (s.d.) 0.08 (0.30) 0.02 (0.16) o (0.40) 0.02 (0.15) o0.001 Ferritin, ng/ml, mean (s.d.) (390.3) (283.3) o0.001 Transferrin saturation index (%) 27.3 (14.3) 28.1 (17.4) C-reactive protein, pg per 100 ml 9.7 (22.5) 7.7 (19.3) (28.0) 7.2 (18.4) o0.001 EPO, erythropoietin; ERI, resistance to ESA index; ESA, erythropoiesis-stimulating agent; Hb, hemoglobin. a Any Hb value out of ±1.5 g per 100 ml of the median Hb level in the total follow-up period for the individual patient. b Any Hb value o11 g per 100 ml. occurred in 20.8% of Hb measurements. Factors significantly associated with Hb variation events and anemia events (Hb levels o11 g per 100 ml) are shown in Table 3. The use of EPO vs long-acting darbepoetin, change of ESA dose in the previous visit, ESA dose, resistance to ESA index, and previous hospital admission were significantly more frequent in the visits in which Hb variation events and anemia events were recorded compared with the visits in which Hb levels remained stable and visits in which Hb levels 411 g per 100 ml were found. Serum ferritin and C-reactive protein levels were also significantly higher in the visits in which an anemia event was recorded compared with visits without an anemia event. In the logistic regression analysis, treatment with EPO, hospital admission, resistance to ESA index, and change of ESA dose in the previous visit were independent predictors of both Hb variation events and anemia events (Table 4). In the Kaplan Meier analysis, median time to the first Hb variation event was 3 months (range months). The median time to the first anemia event was also 3 months (range months) (Figure 1). Table 4 Results of multivariate analysis: risk factors for the appearance of Hb variation events and anemia events at each monthly visit Odds ratio (95% confidence interval) Hb variation events a Hospital admission (yes/no) 3.10 ( ) Long-acting darbepoetin vs EPO 0.70 ( ) Long-acting darbepoetin vs EPO, adjusted by 0.74 ( ) center, age, and type of vascular access Maintenance of ESA dose vs change 0.58 ( ) ERI, U/week/g Hb 1.01 ( ) Anemia events b Hospital admission (yes/no) 9.76 ( ) Long-acting darbepoetin vs EPO 0.72 ( ) Long-acting darbepoetin vs EPO, adjusted by 0.70 ( ) center, age, and type of vascular access Change of ESA dose vs maintenance 3.35 ( ) ERI, U/week/g Hb 1.01 ( ) EPO, erythropoietin; ERI, resistance to ESA index; ESA, erythropoiesis-stimulating agent; Hb, hemoglobin. a Any Hb value out of ±1.5 g per 100 ml of the median Hb level in the total follow-up period for the individual patient. b Any Hb value o11 g per 100 ml. S84

4 JM Portolés et al.: Maintaining hemoglobin target in hemodialysis patients o r i g i n a l a r t i c l e Probability Time Anemia event Hb variation event Figure 1 Survival curve (Kaplan Meier method) until the first anemia event (Hb o11 g per 100 ml) (continuous line) and the first Hb variation event (±1.5 g per 100 ml of the median Hb level in the total follow-up period for the individual patients) (dotted line). Hb, hemoglobin. DISCUSSION Different studies carried out in patients with end-stage renal disease demonstrate that Hb values lower than 10 g per 100 ml have a negative impact on the quality of life and are associated with an increase in morbidity and mortality in CKD patients. 1,15 However, the upper Hb target recommended for these patients remains controversial. Four randomized controlled studies have compared full and partial anemia correction in ESA-treated hemodialysis patients. In a study of patients with clinical evidence of congestive heart failure or ischemic disease assigned to receive increasing doses of EPO to achieve and maintain a hematocrit of 42% (normal hematocrit group) or 30% (low hematocrit group), the study was halted because of an increase in mortality in the normal hematocrit group. 16 The Canadian European study failed to demonstrate a beneficial effect on the left ventricular size and other functional parameters of a higher Hb target ( g per 100 ml) compared with a lower target ( g per 100 ml). 17 Two randomized trials published in 2006 and conducted in non-dialysis patients with CKD carried out in the United States 6 and Europe 7 also showed that complete correction of anemia was not associated with a significant reduction of cardiovascular events, and only one of them showed an improvement in quality of life in the high-hb group. 7 One year later, the National Kidney Foundation (KDOQI) clinical practice guidelines for anemia in CKD recommended that the Hb target should be maintained in the range of g per 100 ml and Hb levels should not exceed 13 g per 100 ml. 8 However, in this study, only 3.8% of the patients maintained target Hb levels between 11 and 13 g per 100 ml during the 1-year follow-up period. This finding is particularly relevant, given that a group of patients with Hb stability before enrollment was selected, which indicates that it is almost impossible to maintain the currently recommended target. Similarly, in the study by Ebben et al., 18 only 10.3% of patients maintained stable Hb levels during 6 months and only 6.5% in the target range. 8 Fluctuation of Hb levels over time in ESA-treated hemodialysis patients, observed by many authors, 10,11,18,19 is likely a consequence of different factors. The interassay variability (coefficient of variation) of Hb measurements is very low and accounts for a minimum part of Hb variability. Although cyclic patterns of Hb fluctuation have been previously reported, 18 the phenomenon of Hb cycling was not observed in our study, in which clinical events such as hospitalization, bleeding or surgical operations, or ESA dose adjustments were identified as risk factors for Hb variation events. In this respect, it should be noted that in two-third of cases, dose changes were greater than 25% and higher than those recommended in the guidelines. 8,20 Fishbane and Berns 10 also found that changes in recombinant human EPO dose were the most important factor associated with up Hb excursions in approximately 80% of cases. Computerbased treatment algorithms to guide ESA dosing have shown to enhance efficiency in the management of renal anemia 21,22 but have not been widely implemented in clinical practice. In this study, the long-acting ESA darbepoetin achieved better Hb stability than short-acting EPO. Although our data were obtained in an observational study, the significant independent effect of darbepoetin persisted after adjustment by variables asymmetrically distributed in both ESA treatment groups, such as center, age, and the type of vascular access. Pharmacokinetic and pharmacodynamic differences between short-acting EPO and long-acting darbepoetin may explain this difference, which may be even more notorious with the use of new ESA methoxypolyethyleneglycol EPO-b (continuous erythropoiesis-stimulating agent) with a longer half-life. Data on the impact of type of ESA on Hb variability in hemodialysis patients are controversial. In 99 hemodialysis patients, whose ESA remained unchanged for a year, Pile et al. 23 found that the type of ESA had no overall effect on the frequency of Hb excursions. No differences in intrapatient variability were observed in a post hoc analysis of a phase III clinical trial comparing subcutaneous continuous erythropoiesis-stimulating activity once monthly or twice monthly with EPO one or three times a week. 24 However, no clinical trials have been conducted with the specific objective of addressing differences in Hb variability associated with the type of ESA treatment. In the overall analysis of the study cohort, it seems that mean Hb values remain stable. However, the analysis of individual patients indicates that it is not possible to maintain Hb within the recommended target range for more than 3 months even in hemodialysis patients with previously stable Hb levels. Clinical outcome is influenced by maintenance of the Hb values on the target range, but retrospective studies have also shown that outcome is related to Hb variability itself. In a retrospective cohort study using data of 41,919 stage 5 CKD patients, the amount of time a patient spent with Hb o11 g per 100 ml increased the risk of death and hospitalization. 25 Gilbertson et al. 26 also found that the number of months with Hb values below the target range may be the primary driver of increased risk of death. In S85

5 o r i g i n a l a r t i c l e JM Portolés et al.: Maintaining hemoglobin target in hemodialysis patients a retrospective cohort of 34,963 hemodialysis patients, survival analysis indicated that each 1 g per 100 ml increase in the residual standard deviation was associated with a 33% increase in rate of death. 27 Consequently, maintaining Hb levels within the KDOQI-recommended range remains difficult but, on the other hand, maintaining Hb levels out of the target appears to negatively affect clinical outcome. In fact, the current guidelines recommend that the objectives should be individualized and to keep stable Hb values within a narrow range. It is unclear whether Hb variability by itself is harmful or simply identifies patients with higher comorbidity or major clinical events and, therefore, with a poorer prognosis. However, achieving Hb stability seems to be a goal independently of other considerations because of its favorable impact on the patient s prognosis Accordingly, the objectives of the treatment of renal anemia should not only be to reach a determined mean Hb level for a group of patients, but also to decide which the appropriate range is for a given patient, how rapidly this target should be achieved, and how to maintain Hb within target range as long as possible and with the lowest consumption of ESA. In conclusion, in a prospective cohort of ESA-treated hemodialysis patients with stable Hb levels before enrollment, individual data indicate that it is extremely difficult to maintain a very narrow target currently recommended by the anemia guidelines, as only 3.8% of patients had Hb levels within target over 1 year. Hb variation events were associated with clinical events or changes of ESA doses, and occurred less frequently in patients treated with long-acting darbepoetin, although these findings need to be confirmed in studies especially designed for this purpose. MATERIALS AND METHODS A historical prospective, cohort, multicenter study was designed to assess maintenance of a target Hb range (11 13 g per 100 ml) in ESAtreated hemodialysis patients and to identify risk factors for Hb variation events. The duration of the study was 1 year. We enrolled patients with end-stage renal disease receiving dialysis between 2004 and 2006 at seven medical centers in Spain attending a reference population of 2 million inhabitants. Hemodialysis patients with Hb stability (within the target range) during the maintenance phase of ESA treatment were included in the study. Hb stability at inclusion was defined by three consecutive monthly values of Hb concentration within the range of g per 100 ml. Patients within the target Hb levels and without data of iron deficiency, hospital admission, transfusion, and major clinical events during this 3-month period were included in the study cohort and followed prospectively for 1 year. In addition, a minimum of 11 monthly laboratory determinations (biochemical and hematological analyses) during the 1-year follow-up period was required. The study protocol was approved by the institutional review board of the participating hospitals. Information was collected by nephrologists in specially designed case report form notebooks. Researchers entered information on demographic characteristics (sex and age), type of center, etiology of CKD, time on dialysis, type of vascular access, and comorbid conditions. Data regarding treatment of anemia (ESA and iron administration) were monthly recorded as well as results of laboratory tests (Hb, serum iron, ferritin, and transferrin saturation index), nutrition and inflammatory markers (serum albumin, highsensitivity C-reactive protein), and other concomitant treatments. Dates of admission to the hospital and blood transfusion administration were registered and added to previously recorded events data at each visit. Resistance to ESA index was defined as the weekly EPO dose (IU/kg) divided by Hb level (g per 100 ml). A ratio of 1:200 was used to convert darbepoetin-a to the EPO equivalent dose (1 mg of darbepoetin-a ¼ 200 IU of epoetin-a or -b). Fulfilment of the treatment goals was Hb g per 100 ml, transferrin saturation index 420%, and ferritin 4100 ng/ml based on the European Best Practice Guidelines on anemia management. 20 Four groups were defined on the basis of Hb levels during the entire follow-up period: compliant patients (all Hb levels within the target), compliant patients with increases (all Hb levels 411 g per 100 ml and some values 413 g per 100 ml), compliant patients with decreases (all Hb levels o13 g per 100 ml and some values o11 g per 100 ml), and patients with Hb instability (some values o11 and 413 g per 100 ml). Anemia events (Hb levels o11 g per 100 ml), Hb variation events (any value out of ±1.5 g per 100 ml of the median Hb level in the total follow-up period for the individual patient), risk factors for anemia and Hb variation events, and time to anemia and Hb variation events were assessed. Statistical analysis Comparison between groups was performed using the w 2 test for categorical variables and the Student s t-test for continuous variables. Differences between compliant patient groups were analyzed with the one-way analysis of variance. Logistic regression analysis was used to identify factors associated with Hb variation events and anemia events. The Kaplan Meier method was used to assess time to first Hb variation event and anemia event. Analysis was performed using SPSS version 15.0 for Windows (SPSS Inc., Chicago, IL, USA). DISCLOSURE Jóse M Portolés, Jóse L Górriz, Alberto Martínez-Castelao, and Evaristo Fernández have done consultancy for Janssen-Cilag, Amgen and Roche; Ángel LM de Francisco and Aleix Cases have done consultancy for Amgen and Roche; Pedro Aljama has done research for Amgen and Roche; Juan M López-Gómez has received consulting and lecture fees from Roche, Amgen and Janssen Cilag. ACKNOWLEDGMENTS This study was carried out under the auspices of the Anemia Study Group of the Spanish Society of Nephrology and the research net REDinREN (Instituto de Salud Carlos III, Red 6/0016). We thank Marta Pulido, MD, for editing the paper and for editorial assistance. REFERENCES 1. Locatelli F, Pisoni RL, Combe C et al. Anaemia in haemodialysis patients of five European countries: association with morbidity and mortality in the Dialysis Outcomes and Practice Patterns Study (DOPPS). Nephrol Dial Transplant 2004; 19: Perlman RL, Finkelstein FO, Liu L et al. Quality of life in chronic kidney disease (CKD): a cross-sectional analysis in the Renal Research Institute- CKD study. Am J Kidney Dis 2005; 45: Ma JZ, Ebben J, Xia H et al. Hematocrit level and associated mortality in hemodialysis patients. J Am Soc Nephrol 1999; 10: Paganini EP. Overview of anemia associated with chronic renal disease: primary and secondary mechanisms. Semin Nephrol 1989; 9(Suppl 1): Phrommintikul A, Haas SJ, Elsik M et al. Mortality and target haemoglobin concentrations in anaemic patients with chronic kidney disease treated with erythropoietin: a meta-analysis. Lancet 2007; 369: S86

6 JM Portolés et al.: Maintaining hemoglobin target in hemodialysis patients o r i g i n a l a r t i c l e 6. Singh AK, Szczech L, Tang KL et al. Correction of anemia with epoetin alfa in chronic kidney disease. N Engl J Med 2006; 355: Drüeke TB, Locatelli F, Clyne N et al. Normalization of hemoglobin level in patients with chronic kidney disease and anemia. N Engl J Med 2006; 355: KDOQI, National Kidney Foundation. KDOQI Clinical Practice Guideline and Clinical Practice Recommendations for anemia in chronic kidney disease: 2007 update of hemoglobin target. Am J Kidney Dis 2007; 50: Levin A. Understanding recent haemoglobin trials in CKD: methods and lesson learned from CREATE and CHOIR. Nephrol Dial Transplant 2007; 22: Fishbane S, Berns JS. Hemoglobin cycling in hemodialysis patients treated with recombinant human erythropoietin. Kidney Int 2006; 68: Lacson Jr E, Ofsthun N, Lazarus JM. Effect of variability in anemia management on hemoglobin outcomes in ESRD. Am J Kidney Dis 2003; 41: Portolés J, Lopez-Gomez JM, Aljama P. Anemia management and treatment response in patients on hemodialysis: the MAR study. J Nephrol 2006; 19: De Nicola L, Conte G, Chiodini P et al. Stability of target hemoglobin levels during the first year of epoetin treatment in patients with chronic kidney disease. Clin J Am Soc Nephrol 2007; 2: Chan KE, Lafayette RA, Whittemore AS et al. Facility factors dominate the ability to achieve target hemoglobin levels in hemodialysis patients. Nephrol Dial Transplant 2008; 23: Portolés J, López-Gómez JM, Aljama P. A prospective multicentre study of the role of anaemia as a risk factor in haemodialysis patients: the MAR Study. Nephrol Dial Transplant 2007; 22: Besarab A, Bolton WK, Browne JK et al. The effects of normal as compared with low hematocrit values in patients with cardiac disease who are receiving hemodialysis and epoetin. NEnglJMed1998; 339: Parfrey PS, Foley RN, Wittreich BH et al. Double-blind comparison of full and partial anemia correction in incident hemodialysis patients without symptomatic heart disease. J Am Soc Nephrol 2005; 16: Ebben JP, Gilbertson DT, Foley RN et al. Hemoglobin level variability: associations with comorbidity, intercurrent events, and hospitalizations. Clin J Am Soc Nephrol 2006; 1: Berns JS, Elzein H, Lynn RI et al. Hemoglobin variability in epoetin-treated hemodialysis patients. Kidney Int 2003; 64: Locatelli F, Aljama P, Bárány P et al. Revised European best practice guidelines for the management of anaemia in patients with chronic renal failure. Nephrol Dial Transplant 2004; 19(Suppl 2): ii1 ii Will EJ, Richardson D, Tolman C et al. Development and exploitation of a clinical decision support system for the management of renal anaemia. Nephrol Dial Transplant 2007; 22(Suppl 4): iv31 iv Gaweda AE, Jacobs AA, Aronoff GR et al. Model predictive control of erythropoietin administration in the anemia of ESRD. Am J Kidney Dis 2008; 51: Pile T, McCafferty K, Byrne CJ et al. The impact of type of erythropoiesis stimulating agent on hemoglobin variability in haemodialysis patients [SU-PO546]. Am Soc Nephrol Sulowicz W, Locatelli F, Ryckelynck JP et al. Once-monthly subcutaneous C.E.R.A. maintains stable hemoglobin control in patients with chronic kidney disease on dialysis and converted directly from epoetin one to three times weekly. Clin J Am Soc Nephrol 2007; 2: Ofsthun NJ, LaBrecque J, Keen M et al. The association of mortality and hospitalization with hemoglobin (HB) and missed dialysis treatments in stage 5 chronic kidney disease (CKD) patients with and without cardiac comorbidities. Abstract and poster presented at the KLII Congress of the European Renal Association-European Dialysis Transplantation Association, Istanbul, Turkey, June Gilbertson DT, Ebben JP, Foley RN et al. Hemoglobin level variability: associations with mortality. Clin J Am Soc Nephrol 2008; 3: Yang W, Israni RK, Brunelli SM et al. Hemoglobin variability and mortality in ESRD. J Am Soc Nephrol 2007; 18: Regidor DL, Kopple JD, Kovesdy CP et al. Associations between changes in hemoglobin and administered erythropoiesis-stimulating agent and survival in hemodialysis patients. J Am Soc Nephrol 2006; 17: Brunelli SM, Joffe MM, Israni RK et al. History-adjusted marginal structural analysis of the association between hemoglobin variability and mortality among chronic hemodialysis patients. Clin J Am Soc Nephrol 2008; 3: S87