Several laboratory tests are commonly used to screen for or establish a diagnosis of chronic iron deficiency. The sequence in which abnormal test results appear is given in the box below. In a normal adult on a normal diet made iron deficient by repeated phlebotomy (for experimental reasons), it takes about 3 months before significant anemia (Hb more than 2 gm/dl below normal) appears. The first laboratory indication of iron deficiency is lack of marrow iron on bone marrow aspiration. The next test to become abnormal is the serum iron level. When anemia becomes manifest, it is moderately hypochromic but only slightly microcytic; marked hypochromia and microcytosis are relatively late manifestations of iron deficiency. When the anemia is treated, results of these tests return to normal in reverse order. Even with adequate therapy it takes several months before bone marrow iron appears again.

Peripheral blood smear. The peripheral blood smear in chronic iron deficiency anemia typically shows RBC hypochromia. There is also microcytosis, but lesser degrees of microcytosis are more difficult to recognize than hypochromia. Some of the peripheral blood changes may appear before actual anemia. However, examination of the peripheral blood smear cannot be depended on to detect iron deficiency, since changes suggestive of chronic iron deficiency either may not be present or may be missed. In one study the peripheral blood smear did not show typical RBC changes in as many as 50% of patients with chronic iron deficiency anemia. This happens more often in patients with mild anemia. Even if hypochromia is present, iron deficiency must be differentiated from other conditions that also may produce hypochromic RBC. In severe anemia there is anisocytosis and poikilocytosis in addition to microcytosis and rather marked hypochromia. The anisocytosis means that the RBCs may not all be microcytes. The microcytes of iron deficiency must be differentiated from spherocytes; such distinction is usually not difficult, since in chronic iron deficiency even the microcytes are hypochromic. Bone marrow aspiration reveals mild erythroid hyperplasia and no marrow iron (using iron stains).

Hypochromic anemias

Table 3-1 Hypochromic anemias

Sequence of Test Abnormalities in the Evolution of Chronic Iron Deficiency

Negative iron balance
Decreased bone marrow hemosiderin
Decreased serum ferritin

Increased RBC protoporphyrin levels
Increased total iron-binding capacity
Decreased serum iron

RBC microcytosis
RBC hypochromia

Red blood cell indices. The mean corpuscular volume (MCV) typically is decreased below reference range lower limits, and the RBC distribution width (RDW) is increased by the time iron deficiency anemia has appeared. However, the few studies available indicate that the MCV is normal in about 30%5% (range 24%-55%) of patients. The mean corpuscular hemoglobin (MCH) value is normal in about 20%. There is considerable disagreement on MCH concentration (MCHC) values, with a decrease reported in 21%-81% of patients. Although various factors could have biased these studies, it is probable that chronic iron deficiency will not be detected by RBC indices in a significant number of patients with chronic iron deficiency anemia. Some of these patients, but not all, may have other superimposed conditions that mask the morphologic effects of iron deficiency. Even if the MCV is decreased, iron deficiency must be differentiated from various other conditions that also produce microcytosis.

Reticulocytes. The reticulocyte count is normal in uncomplicated chronic iron deficiency anemia. Superimposed acute blood loss or other factors, such as adequate iron in the hospital diet, may cause reticulocytosis. For a short time following recent (acute) hemorrhage, the Wintrobe MCV may be normal or even increased due to the reticulocytosis. The reticulocyte response to iron therapy (3%-7%) is somewhat less than that seen with treatment of megaloblastic anemia.

Serum iron. Serum iron levels fall sometime between depletion of tissue iron stores and development of anemia. Therefore, the serum iron value should be a sensitive indicator of possible iron deficiency by the time a patient has anemia. Unfortunately, about 10%-15% (literature range 0%2%) of serum iron measurements in patients with iron deficiency anemia remain in the lower half of the reference range.

CONDITIONS THAT AFFECT SERUM IRON LEVELS. The first is transferrin levels. Serum iron measurement predominantly reflects iron bound to serum proteins. Under usual conditions, most iron is bound to transferrin. Normally, transferrin is about one-third saturated. Therefore, serum iron values depend not only on the quantity of iron available but also on the amount of transferrin present. (If transferrin is increased, the serum iron measurement reflects not only the quantity of iron bound to the normal amount of protein but also the iron bound to the additional protein. The opposite happens when transferrin is decreased.) Second is the time of day. There is a 20%0% diurnal variation in serum iron levels (literature range 2%-69%); the time of day at which the peak value appears is most often in the morning, but it may occur in the early or late afternoon. In one study the peak was found at 8 A in 72% of 25 patients and at 4 P in 28%. Therefore, in some patients the time of day that the specimen is obtained can materially influence whether a result is interpreted as mildly decreased or still within the lower reference range. Third, it has also been found that serum iron displays considerable day-to-day variation among individuals, with changes averaging 20%0% but in some cases varying over 100%. Finally, in some cases there may be some degree of iron contamination of laboratory materials.

SERUM IRON DECREASE IN VARIOUS CONDITIONS. Serum iron levels may be decreased in other conditions besides iron deficiency; the most frequent is probably the anemia associated with severe chronic disease such as the rheumatoid-collagen diseases, extensive malignancy, uremia, cirrhosis, and severe chronic infection (Table 3-1). There is usually a slight increase in serum iron levels in the first trimester of pregnancy, since increased estrogens tend to increase transferrin. However, by the third trimester the effect of estrogens is reversed, partially by hemodilution but also from utilization of maternal iron by the fetus. This leads to a decrease in serum iron in the third trimester. Severe stress (surgery, infection, myocardial or cerebral infarction) frequently produces a considerable decrease in serum iron (in one study by an average of 65% with a range of 38%-93%), which begins within 24 hours of the onset of the stress (sometimes as early as 4-6 hours). Its nadir occurs between 24 and 48 hours, and recovery begins toward baseline about 6-7 days after the original decrease.

SERUM IRON INCREASE. Serum iron levels may be increased in hemolytic anemia, iron overload conditions, estrogen therapy (due to an increase in transferrin), acute hepatitis, and parenteral iron therapy. The effects of intramuscular iron-dextran (Imferon) administration persist for several weeks. The serum iron level is normal or increased in thalassemia minor without coexisting iron deficiency(Table 3-2 and Table 37-2).

Serum iron not total iron-binding capacity patterns

Table 3-2 Serum iron not total iron-binding capacity patterns

SERUM IRON IN MEGALOBLASTIC ANEMIA. When megaloblastic anemia is treated, the serum iron level temporarily falls resulting from marked utilization of previously unused available iron. On the other hand, a significant minority of patients with megaloblastic anemia (20%-40%) have coexisting iron deficiency that eventually will be unmasked by correction of the folate or B12 deficiency. Since megaloblastic anemia can interfere with interpretation of tests for iron deficiency, it has been recommended that follow-up studies be done 1 months after the beginning of folate or B12 therapy to rule out iron deficiency.

Serum total iron-binding capacity. Serum total iron-binding capacity (TIBC) is an approximate estimate of serum transferrin. Assay is usually performed by adding an excess of iron to serum to saturate serum transferrin, removing all iron not bound to protein, and then measuring the serum iron (which is assumed to be mostly bound to transferrin under these conditions). Since transferrin is not the only protein that can bind iron, the TIBC is not an exact measurement of transferrin and tends to be even less representative in cases of iron overload and certain other conditions.

Serum TIBC is increased in uncomplicated chronic iron deficiency, most studies indicating abnormality at the same time as a decrease in serum iron levels or even before. Unfortunately, the TIBC is not elevated above reference limits in 30%-40% (29%-68%) of patients with chronic iron deficiency anemia. In the best-known study published, 69% of iron deficiency anemia patients with low serum iron levels had an elevated TIBC, 11% had a TIBC within reference limits, and an additional 21% had decreased TIBC values. Transferrin is a “negative” acute-phase reaction protein and decreases both with various acute diseases and with severe chronic diseases (the same chronic diseases that decrease serum iron levels). Decrease in transferrin depresses TIBC to low or low-normal levels. Hypoproteinemia and iron overload conditions are also associated with a decreased TIBC. Unfortunately, conditions that decrease TIBC can mask the TIBC elevation of coexisting chronic iron deficiency. Some conditions increase transferrin levels and therefore increase TIBC; these include pregnancy, estrogen therapy, alcoholism, and acute hepatitis (Table 3-2 and Table 37-2).

Transferrin saturation. The textbook pattern of iron tests in chronic iron deficiency shows a decrease in serum iron levels and an increase in TIBC. This will increase the unsaturated binding capacity of transferrin and decrease the percent of transferrin that is bound to iron (percent transferrin saturation, or %TS). A %TS of 15% or less is the classic finding in chronic iron deficiency anemia. The %TS is said to be a more sensitive screening test for chronic iron deficiency than either serum iron levels or the TIBC, since a decreased serum iron level that still remains in the lower end of the reference range plus a TIBC still in the upper end of the TIBC reference range may produce a %TS below 15%. A decrease in %TS is also found in many patients with anemia of chronic disease, so that decreased %TS is not specific for iron deficiency. Also, about 15% (10%4%) of patients with iron deficiency have a %TS greater than 15%, especially in the early stages or when iron deficiency is superimposed on other conditions. The %TS is increased in hemolytic or megaloblastic anemia, sideroblastic anemia, and iron overload states and is normal or increased in thalassemia minor (Table 3-2; a more complete list of conditions that affect TIBC and %TS is included in Table 37-2).

Serum ferritin. Ferritin is the major body iron-storage compound. Routine tissues or bone marrow iron stains, however, detect hemosiderin but not ferritin. Ferritin in serum can be measured by radioassay or enzyme immunoassay. A serum ferritin level decrease accompanies a decrease in tissue ferritin level, which, in turn, closely mirrors decrease of body iron stores in iron deficiency. The decrease in tissue ferritin occurs before changes in serum iron tests, changes in RBC morphology, or anemia. Except for bone marrow iron stains, serum ferritin is currently the most sensitive test available for detection of iron deficiency. The major factors that modify its efficacy as an indicator involve the technical aspects of present-day ferritin immunoassay kits, some of which have less than desirable reproducibility and accuracy at the low end of the reference range. A major reason for this is the fact that the lower edge of the reference range (20-150 ng/ml or µg/L) is not far from zero. Another problem is the extreme difficulty most laboratories have in establishing their own ferritin reference range, since there is no good way to exclude subclinical iron deficiency from the clinically “normal” population without performing bone marrow aspiration. A third problem, partially arising from inadequately validated reference ranges, is disagreement in the literature as to what cutoff level should be used to confirm or exclude iron deficiency. The majority of investigators use 12 ng/ml as the cutoff level (literature range 10-20 ng/ml). A fourth problem (discussed later) is increase in ferritin levels by various conditions that may coexist with iron deficiency.

Ferritin levels at birth are very high and are the same for boys and girls. Ferritin values decrease rapidly by age 3 months and reach their lowest point at about age 9 months. At some time during the teenage years the reference ranges for boys and girls being to diverge somewhat, with the lower limit of the reference range for girls being approximately 10 ng/100 ml lower than that for boys. The upper limit in men tends to increase slowly until old age, whereas the upper limit in women tends to remain relatively stationary until menopause and then slowly increases. The lower limits of reference ranges for both sexes are affected only to a small degree by age. There is approximately a 10%-15% average daily variation in ferritin values in the same individual; about one half the variation is due to fluctuation in serum iron values.

INTERPRETATION OF SERUM FERRITIN RESULTS. A serum ferritin level less than 12 ng/ml is considered almost diagnostic of iron deficiency. Presumably false positive results in the literature based on bone marrow iron stains (displaying decreased serum ferritin levels with bone marrow iron present) range from 0%-4% of cases. False negative results have been reported in 2.6% of bone marrow-proven uncomplicated iron-deficient cases. However, if iron deficiency coexists with a condition that raises the serum ferritin, the ferritin value in a substantial number of patients may be higher than the cutoff value for iron deficiency. Serum ferritin level is decreased to a variable degree during pregnancy; the amount of decrease may be reduced as much as 50% if iron supplements are given.

Many conditions can elevate ferritin levels. Serum ferritin is one of a group of proteins that become elevated in response to acute inflammation, infection, or trauma; elevation begins between 24 and 48 hours, peaks in about 3 days, and lasts 5 days to 5 weeks. In addition, a more sustained increase in ferritin levels may be produced by various chronic diseases (see Table 3-1), including those that decrease serum iron and serum TIBC values. Fortunately, some patients with coexisting chronic disease and iron deficiency still have decreased serum ferritin levels. Ferritin values may also be increased in some patients who have had blood transfusions, in megaloblastic anemia, and in hemolytic anemias. Ferritin is greatly increased in iron overload states such as hemochromatosis and acute iron poisoning. One study reports that about one third of patients with chronic hepatitis virus had elevated serum ferritin and some also had elevated serum iron and TIBC, simulating hemochromatosis.

The serum ferritin level has been used in chronic renal failure to monitor iron status. Because chronic disease raises serum ferritin levels, the ferritin lower limit used for this purpose (approximately 100 ng/ml) is much higher than the lower limit of reference range used for the general population.

Free erythrocyte protoporphyrin (zinc protoporphyrin). This test is discussed in detail in Chapter 35. The last step in heme synthesis occurs when the heme precursor protoporphyrin IX forms a complex with an iron atom with the help of the enzyme ferrochelatase (see Fig. 34-1). If iron is not available, or if ferrochelatase is inhibited (as occurs in lead poisoning), a zinc ion becomes complexed with protoporphyrin IX (zinc protoporphyrin; ZPP) instead of iron. When ZPP is assayed using manual biochemical techniques, the zinc ion is removed during acid extraction of RBC hemoglobin, and the metal-free substance measured is then called free erythrocyte protoporphyrin. Zinc protoporphyrin can be measured directly and quickly using one or two drops of whole blood by means of a small commercially available instrument called a hematofluorometer.

ZINC PROTOPORPHYRIN IN IRON DEFICIENCY. Zinc protoporphyrin levels are elevated in iron deficiency and in lead poisoning. In iron deficiency, ZPP levels become elevated after several weeks of deficient iron stores and return to normal only after 2 to 3 months of iron therapy. In two studies, elevated ZPP levels detected 83%-94% of patients who were iron deficient on the basis of low serum ferritin levels.

PROBLEMS WITH ZINC PROTOPORPHYRIN ASSAYS. Some hematofluorometers report ZPP per unit of whole blood; this reporting system may be affected by changes in hematocrit values. This problem is avoided with instruments that report results as a ZPP/heme ratio. Another potential difficulty is falsely decreased results due to a shift in the protoporphyrin fluorescent maximal absorption peak if the Hb is not fully oxygenated. This can be avoided in several ways. More troublesome is ZPP elevation by acute or chronic infections, noninfectious inflammation, various malignancies, chronic liver disease, and moderate or severe hemolytic anemias. Therefore, ZPP levels are elevated in many of the same conditions that falsely elevate serum ferritin levels. Although ZPP is a good screening method for iron deficiency and lead poisoning, most laboratories do not own a hematofluorometer.

Bone marrow iron stain. The gold standard for chronic iron deficiency has been bone marrow aspiration or biopsy with Prussian blue chemical reaction for iron (hemosiderin). Although there is some disagreement, a clot section is generally considered more reliable for iron staining than an aspiration smear. Bone biopsy specimens must be decalcified, and some decalcifying reagents (but not others) may destroy some iron. The major problem with bone marrow aspiration has been reluctance of patients to undergo the procedure. Occasionally, bone marrow aspiration may be necessary to diagnose patients with hypochromic anemia without clear-cut evidence from other tests for or against iron deficiency. However, a therapeutic trial of iron might provide the same information.