INTRODUCTION — Anemia is common among patients with chronic kidney disease (CKD). Anemia underlies many of the symptoms associated with reduced kidney function and is associated with increased mortality and hospitalizations [1-4].
The screening and treatment of anemia in nondialysis CKD patients are discussed here. Our approach to screening and treating anemia among dialysis patients is discussed elsewhere:
●(See "Treatment of iron deficiency in dialysis patients".)
The treatment of iron deficiency among nondialysis CKD patients and dialysis patients is discussed elsewhere:
●(See "Treatment of iron deficiency in nondialysis chronic kidney disease (CKD) patients".)
●(See "Treatment of iron deficiency in dialysis patients".)
DEFINITION — Anemia is defined by the World Health Organization (WHO) as a hemoglobin (Hb) concentration <13.0 g/dL for adult males and postmenopausal women and an Hb <12.0 g/dL for premenopausal women [5]. (See "Diagnostic approach to anemia in adults", section on 'Caveats for normal ranges'.)
However, the WHO definition of anemia does not define goals of treatment among CKD patients. Thus, even when adequately treated, many CKD patients will have anemia as defined above. This is because the treatment of anemia involves erythropoiesis-stimulating agents (ESAs); multiple studies have now shown that targeting a normal Hb with ESAs increases risks of adverse outcomes. (See 'Target hemoglobin value' below.)
EPIDEMIOLOGY — Anemia is common among nondialysis CKD patients, and the prevalence increases as the glomerular filtration rate (GFR) declines [6-8]. As an example, based upon over 15,000 participants in the National Health and Nutrition Examination Survey (NHANES), the prevalence of anemia (hemoglobin [Hb] <12 g/dL in men and <11 g/dL in women) increased from 1 percent among patients with an estimated GFR (eGFR) of 60 mL/min/1.73 m2 to 9 percent at an eGFR of 30 mL/min/1.73 m2 and to 33 to 67 percent at an eGFR of 15 mL/min/1.73 m2 [8].
SCREENING
Initial screening — All patients should be screened for anemia when they are first evaluated for CKD with a complete blood count (CBC).
Patients who are found to be anemic should be evaluated for the cause. Patients may have anemia related to CKD but are also at risk for all the other causes of anemia that occur in the general population. The initial evaluation of anemia is generally the same for CKD patients as in the general population. The evaluation should include CBC, red blood cell (RBC) indices, reticulocyte count, serum iron, total iron-binding capacity (TIBC), percent transferrin saturation (TSAT), serum ferritin, serum folate and vitamin B12 levels, and testing for occult blood in stool. (See "Diagnostic approach to anemia in adults".)
Continued monitoring — After the initial screen and evaluation, we continue to routinely monitor all CKD patients for anemia and, if anemia is present, for iron deficiency. We monitor for anemia with hemoglobin (Hb) concentration. Screening tests for iron deficiency are discussed elsewhere. (See "Treatment of iron deficiency in nondialysis chronic kidney disease (CKD) patients", section on 'Screening for anemia and iron deficiency'.)
The frequency of monitoring varies based upon the presence of anemia on initial evaluation, trends in Hb concentration over time, whether or not patients are treated with erythropoiesis-stimulating agents (ESAs), and upon the severity of CKD. The following approach is largely consistent with the Kidney Disease: Improving Global Outcomes (KDIGO) guidelines [9].
Patients without anemia — For CKD patients without anemia, we monitor as follows:
●Estimated glomerular filtration rate (eGFR) ≥45 mL/min/1.73 m2 – We monitor such patients for development of anemia annually. We evaluate for the cause of anemia as described above if anemia is present. (See "Treatment of iron deficiency in nondialysis chronic kidney disease (CKD) patients", section on 'Initial screen'.)
●eGFR <45 mL/min/1.73 m2 – We screen such patients at least twice yearly for anemia. We evaluate for the cause of anemia as described above if anemia is present. (See "Treatment of iron deficiency in nondialysis chronic kidney disease (CKD) patients", section on 'Initial screen'.)
In addition, among all CKD patients, Hb should be checked whenever clinically indicated (such as after major surgical procedures, hospitalization, or bleeding).
Patients with anemia but not on an ESA — For CKD patients who are anemic on initial screen but are not being treated with an erythropoiesis-stimulating agent (ESA), we monitor the Hb concentration every three to six months depending on the eGFR, Hb concentration, and prior change in Hb concentration:
●eGFR ≥45 mL/min/1.73 m2 – Patients with eGFR 45 to 60 mL/min/1.73 m2 who have mild anemia are monitored every six months. Patients who have had a progressive decrease in Hb or have moderate or severe anemia (ie, Hb <10 g/dL) are monitored every three months.
●eGFR <45 mL/min/1.73 m2 – Patients with eGFR <45 mL/min/1.73 m2 are generally monitored every three months.
Iron studies should be similarly monitored if iron deficiency is present at initial evaluation. Hb and iron studies are also generally measured after blood loss or surgery. (See "Treatment of iron deficiency in nondialysis chronic kidney disease (CKD) patients", section on 'Continued monitoring'.)
Patients treated with an ESA — Among patients who are started on an erythropoiesis-stimulating agent (ESA) or in whom the dose has been increased, we measure the Hb every two to four weeks while the ESA is being initiated and at least every three months thereafter. However, recommendations from the US Food and Drug Administration (FDA) are for more frequent testing.
We generally screen such patients for iron deficiency every three months. Patients who have marginal iron status, or those with declining serum ferritin or TSAT levels on initial screen, should be screened every two to three months. (See "Treatment of iron deficiency in nondialysis chronic kidney disease (CKD) patients", section on 'Continued monitoring'.)
The 2012 KDIGO and National Institute for Health and Care Excellence (NICE) guidelines suggest evaluating iron status at least every three months during ESA treatment and more frequently when increasing the ESA dose or monitoring the response to intravenous iron or when there is blood loss or other circumstances (eg, hospitalization) when iron stores may be depleted [10,11].
There are no data to support monitoring with such frequency, and some clinicians monitor CKD patients less frequently [12,13].
Patients treated with iron — Whether on an ESA or not, we typically assess iron status every three months for patients receiving oral iron. Patients who receive intravenous iron should have iron stores checked after a course of treatment. (See "Treatment of iron deficiency in nondialysis chronic kidney disease (CKD) patients", section on 'Goals of therapy'.)
There are no data to support monitoring with such frequency, and some clinicians monitor CKD patients less frequently [12,13].
TREATMENT — The primary therapeutic options for the anemia of CKD include iron, erythropoiesis-stimulating agents (ESAs), and, rarely, red blood cell (RBC) transfusions. The treatment depends on severity of anemia and iron deficiency.
Iron — Anemic patients who are iron deficient should be treated with iron before the administration of ESAs. (See "Treatment of iron deficiency in nondialysis chronic kidney disease (CKD) patients", section on 'Indications for treatment' and "Treatment of iron deficiency in nondialysis chronic kidney disease (CKD) patients", section on 'Treatment'.)
Erythropoiesis-stimulating agents
Indications and contraindications — We administer ESAs to most CKD patients who have a hemoglobin (Hb) <10 g/dL, providing the transferrin saturation (TSAT) is >20 percent and ferritin >200 ng/mL. Important exceptions are among:
●Patients who have an active malignancy (particularly those in whom cure is anticipated) or a recent history of malignancy. We avoid use of ESAs among such patients because of the possible increase in the risk of progression or recurrence of cancer.
●Patients who have had a stroke. We avoid use of ESAs in such patients since they may be at a higher risk for adverse effects (eg, recurrent stroke) from ESAs.
●Patients who have certain comorbidities (eg, being bedbound or with very limited functional capacity, dementia, etc) that make them unlikely to derive the same benefit from ESAs as someone who is more active and symptomatic from anemia. We avoid use of ESAs in such patients unless there is a need to improve Hb sufficiently to achieve a specific clinical goal (eg, to minimize hospitalizations for transfusions, optimize management of heart failure) that cannot be achieved with iron supplementation alone.
Among patients with TSAT ≤20 percent and ferritin ≤500 ng/mL, we usually administer iron before giving an ESA since they may respond to iron with an increase in Hb. If the response to iron supplementation is adequate, then we do not use ESA in such patients.
We are attentive to possible symptoms of anemia in younger patients who have CKD with few comorbidities, whose symptoms of anemia may occur at higher Hb levels. For such patients, we may initiate ESAs at Hb levels of 10 g/dL or even higher after discussing potential risks and benefits with each patient.
The administration of ESAs has substantially reduced the need for red cell transfusions (with an attendant decrease in and/or risk for transfusion-related complications). (See "Hyporesponse to erythropoiesis-stimulating agents (ESAs) in chronic kidney disease".)
However, the Hb concentration at which to initiate ESAs is not known with certainty, and the safety of ESAs in treating even severe anemia has not been evaluated in large, placebo-controlled trials. We agree with the 2012 Kidney Disease: Improving Global Outcomes (KDIGO) guidelines that the potential benefits of reducing blood transfusions and anemia-related symptoms should be balanced against the harm in individual patients (such as stroke, vascular access loss, and hypertension) [10].
Route of administration — We recommend that, for CKD patients, an ESA, whether epoetin or darbepoetin, be administered subcutaneously. Studies have shown that the subcutaneous dose of epoetin required to achieve a target Hb is approximately 30 percent less than that required with intravenous administration [14,15]. Intravenous and subcutaneous administration of darbepoetin are of similar efficacy. For nondialysis CKD patients, subcutaneous administration is more convenient for the patient as it allows self-administration. The avoidance of intravenous administration also allows the better preservation of veins for future hemodialysis access.
Dosing — The initial dose for epoetin is approximately 50 to 100 units/kg/week. However, the use of lower doses would also be reasonable, particularly in patients with pretreatment Hb levels near 10 g/dL.
In practice, most patients are dosed by unit dosing (eg, a vial), rather than on a strict unit/kg basis. Thus, we initiate therapy in most patients beginning at 4000 or 10,000 units subcutaneously once weekly or 10,000 to 20,000 units subcutaneously every other week. Weekly or even less frequent dosing regimens have been shown to be effective and safe, although long-term studies are lacking [16-22]. A meta-analysis including seven randomized trials demonstrated no difference in Hb when epoetin was administered every two to four weeks compared with more frequent (weekly) dosing intervals [22].
Darbepoetin is typically initiated with doses of 40 to 100 mcg subcutaneously every two to four weeks. If necessary, subsequent adjustments are made in interval and/or dose.
We suggest that the lowest effective ESA dose be used. Higher ESA doses (primarily epoetin doses greater than 10,000 units per week or equivalent darbepoetin doses) have been associated with increased mortality and cardiovascular events independent of Hb level [23]. (See "Treatment of anemia in patients on dialysis".)
Methoxy polyethylene glycol-epoetin beta has also been approved for use in patients with CKD with a recommended initial starting dose of 0.6 mcg/kg by intravenous or subcutaneous injection every two weeks and monthly dosing subsequently. Epoetin alfa-epbx is the first epoetin "biosimilar" agent approved in the United States; as with other epoetins, the package insert recommends thrice-weekly dosing, which is impractical. Its place in treatment of anemia in patients with CKD not on dialysis remains to be determined. (See "Treatment of anemia in patients on dialysis", section on 'Initial dosing'.)
Target hemoglobin value — For most nondialysis CKD patients who are treated with ESAs, we maintain Hb levels between 10 and 11.5 g/dL using the lowest possible ESA dose. We individualize therapy in some patients who may have improvements in quality of life at Hb ≥11.5 g/dL and will be prepared to accept the possible risks associated with higher Hb targets. We do not target Hb concentration >13 g/dL.
The optimal target Hb level for CKD patients is not well defined. Our target Hb range reflects the results of clinical studies and recognizes that anemia treatment should be individualized. The US FDA boxed warning on ESAs states that Hb targets >11 g/dL are not recommended [24]. While this recommendation is appropriate for most patients, our upper target Hb limit of 11.5 g/dL acknowledges that some patients may have better anemia symptom control and quality of life (eg, improved exercise capacity) with Hb >11 g/dL and is based on Hb concentrations generally not exceeding 11.5 g/dL in the control group of major randomized controlled trials. Our practice is consistent with the KDIGO 2012 guidelines [10].
Among both dialysis and nondialysis CKD patients, multiple studies have shown that targeting Hb levels ≥13 g/dL is associated with increased risk of adverse outcomes [25-28].
A number of randomized trials have compared Hb target levels for predialysis patients with CKD [25,26,29]. The best data are from the Trial to Reduce Cardiovascular Events with Aranesp Therapy (TREAT) trial, in which 4038 patients with type 2 diabetes and CKD (estimated glomerular filtration rate [eGFR] between 20 to 60 mL/min/1.73 m2) were randomly assigned to receive darbepoetin alfa to achieve a target Hb level of 13 g/dL or to placebo, with darbepoetin administered if the Hb level was <9 g/dL [29]. The primary endpoints were the composite outcomes of death or a cardiovascular event (nonfatal myocardial infarction [MI], heart failure, stroke, and hospitalization for myocardial ischemia) and death or end-stage kidney disease (ESKD). The mean achieved Hb level was 12.5 g/dL and 10.6 g/dL in the darbepoetin and placebo groups, respectively.
At a median follow-up of 29 months, both groups had similar risks of death or a cardiovascular event or death or ESKD. However, there was an increased risk of fatal or nonfatal stroke with darbepoetin alfa (101 versus 53 patients with placebo; hazard ratio [HR] 1.92, 95% CI 1.38-2.68), while red cell transfusions were significantly more common in the placebo group (496 versus 297 patients). Fatigue was only modestly less common with darbepoetin. There was also an increased risk of death due to malignancy in the darbepoetin group, primarily in patients with a past history of malignancy.
In the TREAT and other trials comparing lower and higher Hb targets in patients with CKD, the patients were diabetic, and many had preexisting cardiovascular disease, likely contributing to a high rate of adverse outcomes. In another, smaller trial of nondiabetic patients treated with darbepoetin, targeting an Hb of 11 to 13 g/dL, compared with 9 to 11 g/dL, led to similar cardiovascular and overall survival [30]. These data support our practice of targeting Hb levels between 10 and 11.5 g/dL in most patients.
Adverse effects — The adverse effects associated with erythropoietin in predialysis patients are similar to those observed in hemodialysis patients. (See "Treatment of anemia in patients on dialysis", section on 'Adverse effects of erythropoiesis-stimulating agents'.)
An important issue that is shared is the increased risk of adverse cardiovascular effects with increased Hb levels. However, hypertension due to erythropoietin is less of an issue in predialysis patients.
Investigational agents — Hypoxia-inducible factor prolyl hydroxylase inhibitors (HIF PHIs) are a novel class of oral ESAs that have been evaluated for treatment of anemia in nondialysis CKD patients [31-40]. Unlike other ESAs that replace endogenous erythropoietin, HIF PHIs stimulate transcription of the erythropoietin gene in the kidneys and liver, leading to increased levels of endogenous erythropoietin. The following HIF PHIs have been evaluated in clinical trials:
●Daprodustat – The safety and efficacy of daprodustat as compared with darbepoetin was evaluated in a trial of 3872 patients with nondialysis CKD and anemia (median Hb 10 mg/dL) who were randomly assigned to daprodustat (dose range, 1 to 24 mg daily) or darbepoetin; patients were followed for a median of approximately two years [41]. Hemoglobin concentrations increased more with daprodustat therapy (by 0.74 versus 0.66 g/dL). However, cardiovascular events (a composite of death, nonfatal stroke, and nonfatal myocardial infarction) were more frequent with daprodustat. The higher cardiovascular event rate with daprodustat was statistically significant during the period of the trial when patients were receiving randomized treatment (14.1 versus 10.5 percent; HR 1.40, 95% CI 1.17-1.68); however, this difference became nonsignificant by the end of the posttreatment follow-up (19.5 versus 19.2 percent).
●Vadadustat – The safety and efficacy of vadadustat was compared with darbepoetin in two trials, one each for ESA-untreated and ESA-treated patients, with a combined total of 3476 patients [32]. Patients were randomly assigned to receive vadadustat (300 to 600 mg) or darbepoetin to aim for a target Hb of 10 to 11 g/dL in the United States and 10 to 12 g/dL in other countries. Between weeks 24 and 36, the mean change in Hb concentration was comparable between vadadustat and darbepoetin groups in the ESA-untreated (1.43 g/dL and 1.38 g/dL, 95% CI -0.04 to 0.15) and ESA-treated (0.41 g/dL and 0.42 g/dL, 95% CI -0.09 to 0.07) patients. The need for red cell transfusion between weeks 24 and 36 was slightly higher with vadadustat in both trials (2.7 versus 2.2 percent in ESA-untreated and 1.6 versus 1.2 percent in ESA-treated); the need for rescue therapy with an ESA was lower in the vadadustat group. Results were largely unchanged through week 52 in both trials.
However, in a pooled analysis of both trials, those assigned vadadustat had a higher rate of major adverse cardiovascular events (MACE) compared with the darbepoetin group (HR 1.17, 95% CI 1.01-1.36). Of the individual components of this composite endpoint, the vadadustat group had a higher rate of death from any cause (18.3 versus 17.7 percent), nonfatal myocardial infarction (3.9 versus 2.8 percent), and nonfatal stroke (2.0 versus 1.6 percent). Other adverse events were similar between the groups. Similar findings were reported in another small trial [33].
●Roxadustat – The efficacy of roxadustat was examined in a phase 3 trial of over 2700 patients with nondialysis CKD and anemia (baseline mean Hb 9 g/dL) who were randomly assigned to receive roxadustat, or placebo thrice weekly, targeting an Hb of at least 11 g/dL [34]. Compared with those taking placebo, patients taking roxadustat were more likely to achieve the target Hb (77 versus 9 percent) and achieved an Hb in the range of 10 to 12 g/dL for a greater proportion of time (82 versus 28 percent). In addition, compared with placebo, roxadustat led to a 74 percent overall reduction in the need for other treatment of anemia (with iron, ESA, or transfusions) and 63 percent reduction in the need for transfusions alone. However, all-cause mortality (21 versus 18 percent), cardiovascular events (23 versus 21 percent), and overall serious adverse events (57 versus 54 percent) were higher among patients on roxadustat compared with patients on placebo. Serious adverse events that were more common with roxadustat included hypertension, urinary tract infections, and pneumonia. Similar findings have been reported in other smaller phase 3 trials from the European Union [35], China [36], Japan [37], and other countries [38]. The FDA declined approval of roxadustat in the United States due to these safety concerns.
Since HIF pathways regulate or interact with many biologic processes, there is concern about nonerythropoietic adverse effects, including increased risk of cancer, thrombosis, cardiovascular disease, progression of diabetic retinopathy, and CKD, among others, which will require long-term follow-up of treated patients [42]. None of these agents are yet approved for use in the United States.
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Anemia in chronic kidney disease".)
SUMMARY AND RECOMMENDATIONS
●Anemia is common among nondialysis chronic kidney disease (CKD) patients. The prevalence increases as the glomerular filtration rates (GFRs) decline. (See 'Epidemiology' above.)
●All patients are screened for anemia when they are first evaluated for CKD and regularly monitored thereafter. The frequency of monitoring varies based upon the presence of anemia on initial evaluation, whether or not patients are treated with erythropoiesis-stimulating agents (ESAs), and upon the severity of CKD. (See 'Screening' above.)
●The treatment of anemia among nondialysis patients with CKD should be individualized. For most CKD patients, we suggest initiating ESAs when the hemoglobin (Hb) level is <10 g/dL, providing the transferrin saturation (TSAT) is >20 percent and ferritin >200 ng/mL (Grade 2C).
An important exception is among patients with an active malignancy or recent history of malignancy, particularly those in whom cure is anticipated, or who have had a stroke since such patients may be at higher risk for adverse effects from ESAs. (See 'Indications and contraindications' above.)
●For most patients who are selected for ESA treatment, we suggest subcutaneous rather than intravenous ESA administration (Grade 2C). The initial dose of epoetin is approximately 50 to 100 units/kg/week and of darbepoetin is 40 to 100 mcg every two to four weeks. Further titration is based on Hb response. (See 'Route of administration' above and 'Dosing' above.)
●The optimal target Hb level for CKD patients is not well defined. For most patients with CKD who are not on dialysis and are on ESAs, we suggest maintaining Hb levels between 10.0 and 11.5 g/dL (Grade 2C). Some clinicians would allow an Hb level >11.5 g/dL for younger patients with CKD who have few comorbidities and who have persistent, severe symptoms of anemia. There are no data on the benefits of Hb concentrations between 11.5 and 13.0 g/dL. (See 'Target hemoglobin value' above.)
●Among nondialysis CKD patients who are treated with ESAs, we recommend not targeting Hb levels >13 g/dL (Grade 1B). Hb targets >13 g/dL are associated with adverse outcomes. (See 'Target hemoglobin value' above.)
