NKF K/DOQI GUIDELINES 2000
I. Anemia Work-Up
When to Initiate the Work-Up of Anemia
An anemia work-up should be initiated in patients with chronic kidney disease (CKD) when the:
Hgb <11g/dL (Hct is <33%) in pre-menopausal females and pre-pubertal patients (Evidence)
Hgb <12g/dL (Hct is <37%) in adult males and post-menopausal females (Evidence)
Table IV-1. Mean Normal Values of Hemoglobin and Hematocrit for the Healthy, Normal Population47
|Age/Gender||Hemoglobin (g/dL)||Hematocrit (%)|
|Birth||16.5 ± 3.0||51 ± 9|
|1 month||14.0 ± 4.0||43 ± 6|
|2 to 6 months||11.5 ± 2.5||35 ± 7|
|6 months to 2 years||12.0 ± 1.5||36 ± 3|
|2 to 6 years||12.5 ± 1.0||37 ± 3|
|6 to 12 years||13.5 ± 2.0||40 ± 5|
|12 to 18 years (male)||14.5 ± 1.5||43 ± 6|
|Menstruating female||14.0 ± 2.0||41 ± 5|
|Adult male/post-menopausal female||15.5 ± 2.0||47 ± 6|
Rationale Anemia is defined in terms of the Hgb or Hct concentration. In this guideline, we recommend that a work-up of anemia be initiated when the Hgb/ Hct level declines to approximately 80% of the mean level for defined healthy, normal subgroups (see Table IV-1: eg, in females, 80% of Hct 41 = Hct 33; in males, 80% of Hct 47 = Hct 37). Differences in average Hgb/Hct levels between adult men and women are likely due to differences in estrogen and testosterone production that emerge at puberty, but subside after menopause. Anemia is likely to be present in individuals when Hgb/ Hct concentrations are below these levels. However, the mean Hgb/ Hct in the general population is only a statistical benchmark and may not be the best indication of anemia in every individual. For example, there is a 75% likelihood of anemia in an adult female with a Hct of 34% or a Hgb of 11 g/dL, or in a male with a Hct of 39% or a Hgb of 12.5 g/dL. Moreover, many individuals have Hgb/Hct concentrations which are physiologically normal for them, but which would be defined as anemia in terms of the general population data. Others have a Hgb/Hct level that may be physiologically inadequate for them (eg, patients with chronic obstructive pulmonary disease), even though it falls within the range considered normal for the general population.
In hemodialysis patients, blood samples to document and monitor anemia should be obtained prior to or immediately upon initiation of the dialysis procedure (predialysis). While a Hgb/Hct obtained at the end of the dialysis procedure (postdialysis) may relate better to a patients estimated dry weight, experience and data reported in the literature universally refer to predialysis Hct and Hgb levels; hence the need to relate these guidelines to predialysis blood samples.
An automated cell counter should be used to determine RBC indices, Hct, and Hgb because the results are more easily standardized. Automated cell counters also have the advantage of providing a total white blood cell count and, often, a platelet count.
Outside of the United States, Hgb, rather than Hct, is used to quantify the level of anemia in patients with CKD. There are several reasons why Hgb is a more accurate, and hence better measure of anemia than is Hct. First, whereas Hgb is stable when a blood sample is stored at room temperature, Hct is not. Specifically, MCV (from which Hct is calculated: MCV _ erythrocyte count = Hct) is stable at room temperature for only 8 hours and is stable for only 24 hours when a blood sample is refrigerated.41 When a blood sample is stored for longer periods of time, MCV increases, resulting in increases in calculated Hct by as much as 2% to 4%.42 In contrast, Hgb remains unchanged when a blood sample is stored for the same amount of time under the same conditions.42 The sensitivity of Hct to blood sample storage conditions is particularly important in light of increased consolidation in the dialysis industry in the United States and the resulting tendency for dialysis centers that comprise a given dialysis chain to ship blood samples over variable distances, under poorly controlled conditions, to centralized laboratories.
A second reason why Hgb is a more accurate measure than Hct is that in the presence of hyperglycemia, MCV (but not Hgb) is falsely elevated, resulting in a false elevation of calculated Hct.43,44 Finally, there is greater variability across automated analyzers in estimation of the number and size of erythrocytes that are in a blood sample (and hence in calculation of Hct) than there is in measurement of Hgb.45 Data comparing the within-run and between-run coefficient of variation (CV) in automated analyzer measurements have shown that these CVs for measurement of Hgb are one half and one third those for Hct, respectively.46
For all these reasons, Hgb is a better measure to use to monitor and manage anemia in patients with CKD than is Hct, particularly given the growing tendency for dialysis centers to send blood samples to outside laboratories, rather than measuring Hgb or Hct in-house. In addition, use of hemoglobin will allow better comparison of anemia management between countries, since most other countries use measurement of hemoglobin as the standard. Therefore, the Anemia Work Group strongly urges that hemoglobin be the primary means of quantifying the level of anemia in patients with CKD.
A. Evaluation of anemia should consist of measurement of at least the following: (Evidence)
Hemoglobin (Hgb) and/or Hematocrit (Hct)
Red blood cell (RBC) indices
Reticulocyte count Iron parameters:
Total Iron Binding Capacity (TIBC)
Percent transferrin saturation (serum iron × 100 divided by TIBC) [TSAT]
A test for occult blood in stool
B. This work-up should be performed before Epoetin therapy is begun. (Opinion)
Fig IV-1. Anemia work-up for CKD patients. Asterisk indicates that laboratory values are consistent with uncomplicated iron deficiency.
Rationale The red blood cell indices, reticulocyte count (an index of new red blood cell formation), and iron parameters are helpful to detect the cause of many anemias which are not due to EPO deficiency.48 The anemia of CKD is generally normocytic and normochromic. Microcytosis may reflect iron deficiency, aluminum excess, or certain hemoglobinopathies; macrocytosis may be associated with vitamin B12 or folate deficiency. Macrocytosis can also be associated with iron excess49 and/or Epoetin therapy that shifts immature, larger reticulocytes into circulation. An elevated reticulocyte count (corrected for the degree of anemia) suggests that active hemolysis may be present, such as in acute renal failure due to the hemolytic uremic syndrome. An abnormal white blood cell count and/or platelet count may reflect a more generalized disturbance of bone marrow function, such as that due to malignancy or vasculitis.
Iron is critical for Hgb synthesis. Consequently, patients should be carefully evaluated for the availability of iron, by measuring the serum iron and the TIBC. The serum iron and the percent TSAT reflect the amount of iron immediately available for hemoglobin synthesis. The serum ferritin reflects total body iron stores. A low level of either of these indices may indicate the need for supplemental iron to support erythropoiesis. Iron deficiency has been shown to be present in as many as 25% to 37.5% of patients presenting with the anemia of CKD50,51 and, if treated, can at least temporarily improve or correct the anemia52 (Endnote a). Absolute iron deficiency in the general population is indicated by a TSAT of less than 16%53 and/or a serum ferritin value of less than 12 ng/mL.54 However, higher values of TSAT and serum ferritin may be necessary to achieve an erythropoietic response prior to initiation of Epoetin therapy (Endnote a) and higher values for these parameters will be required to support accelerated erythropoiesis stimulated by pharmacological administration of Epoetin (see Guideline 6: Target Iron Level). The presence of iron deficiency requires a search for the cause, which is usually blood loss. A stool guaiac test for occult blood is recommended to test for gastrointestinal bleeding in patients with iron deficiency. Another test for early iron deficiency is an increase in the number of hypochromic red blood cells determined by certain autoanalyzers, ie, Technicon H-1, H-2, and H-3 Autoanalyzers (Bayer Diagnostics). A hypochromic red blood cell is defined as an individual cell with an Hgb concentration of <28 g/dL. Normally, less than 2.5% of red blood cells are hypochromic. Although the autoanalyzer used to perform this test is available in Europe,55,56 it is not readily available in the United States at this time. Because of the limited availability of these autoanalyzers in the United States, this test has not been included as part of the guideline. However, if such technology becomes routinely available in the United States, this test should be considered in the work-up of the anemia of CKD, particularly since Epoetin therapy may increase the likelihood of functional iron deficiency. In CKD patients without iron deficiency, it is prudent to screen for common causes of anemia other than EPO deficiency (see Guideline 3: Erythropoietin Deficiency). Correcting an easily reversible cause of anemia makes both clinical and economic sense. An example is hypothyroidism, which is common in the general population, and can cause a normochromic, normocytic anemia that can mimic the anemia due to EPO deficiency.33 If a reversible cause of anemia is not present or has been corrected, and EPO deficiency is the likely primary cause of the anemia, then anemia should be treated with Epoetin to improve patient quality of life,24 to improve the various physiological abnormalities associated with anemia, to decrease morbidity,37 to decrease hospitalization,57 and to improve patient survival.36
If no cause for anemia other than CKD is detected, based on the work-up outlined in Guideline 2: Anemia Evaluation, and the serum creatinine is Ù2 mg/dL, anemia is most likely due to EPO deficiency. Measurement of serum EPO levels usually is not indicated. Fig IV-1 provides a guideline for the work-up of anemia in patients with a serum creatinine Ù2 mg/dL, and for those occasional patients with a lower serum creatinine and impaired kidney function who have a normocytic, normochromic anemia. (Evidence)
Rationale As kidney function declines, the likelihood of anemia associated with EPO deficiency increases because the diseased kidneys are unable to produce sufficient quantities of EPO. Anemia can develop relatively early in the course of CKD, however, and has been associated with a serum creatinine as low as 2.0 mg/dL,58 and occasionally even lower, particularly in individuals with a reduced muscle mass. On the other hand, there is a wide range of Hgb/Hct levels for any degree of kidney dysfunction. Two studies have found a linear relationship between Hct and creatinine clearance in pediatric patients. A linear relationship between GFR and Hct was observed in 48 pediatric patients in one study,59 and in 31 CKD pediatric patients in another study when the GFR was estimated from the serum creatinine.60 In these two studies, significant anemia was noted when the GFR was less than 20 and 35 mL/min/1.73 m2, respectively.
The anemia of CKD should not be confused with the anemia of chronic disease. In the latter, inflammatory cytokines suppress the endogenous production of EPO and erythropoiesis directly.61,62 Measurable levels of circulating cytokines may be found in stable dialysis patients, but, in the absence of inflammation, do not appear to adversely affect the action of Epoetin63,64 (see Guideline 20: Causes for Inadequate Response to Epoetin).
In patients with non-renal anemia, serum EPO levels are usually elevated in an effort to compensate for the anemia. In patients with impaired kidney function and a normochromic, normocytic anemia, it is rare for the serum EPO level to be elevated. Therefore, measurement of EPO levels in such patients is not likely to guide clinical decision-making or Epoetin therapy.
Figure IV-1 suggests an approach for evaluating anemia in CKD patients who do not have gastrointestinal bleeding.
© 2001 National Kidney Foundation, Inc