Pediatric Annals

Problems in Diagnosis of Iron Deficiency Anemia

Philip Lanzkowsky, MD, FRCP, DCH

Abstract

Iron deficiency anemia is the most common nutritional deficiency in children and is widespread in childhood populations throughout the world. It is especially prevalent in infancy. Many factors make infants and children vulnerable to develop negative iron balance and iron deficiency anemia. These include the fact that their diet often contains only trace amounts of iron, their limited ability to absorb dietary iron, their need of iron for growth, as well as the high prevalence of parasitism and gastrointestinal blood loss in some populations.

Iron, which is abundant in the earth's crust, has unique and subtle chemical properties and carries out a wide range of biological functions. It is critical for certain metabolic and enzymatic processes, is essential for growth, and plays a vital role in the structure of the hemoglobin molecule. It is responsible for the transport of oxygen (as hemoglobin and myoglobin), it participates in the activation of both molecular nitrogen and oxygen (as nitrogenases, oxygenases and oxidases) and plays a role in electron transport (as cytochromes). In addition, because of the propensity of iron to hydrolyse in aqueous solution, special molecules have been designed for its transport (transferrin) and storage (ferritin).

PREVALENCE OF IRON DEFICIENCY

Surveys of preschool children in many countries in the last half century have revealed widespread iron deficiency anemia. The incidence of the disorder has varied from population to population. Incidence is dependent upon a number of factors, such as the age group of the population selected, ethnic composition, dietetic habits, socioeconomic factors, presence of intestinal parasites, and the methods used for the detection of iron deficiency.

A number of surveys carried out in urban areas of the US over the last 20 years have revealed that the incidence of iron deficiency anemia in children between 6 and 36 months of age varies from 17% to 44%. A survey of 417 children from 6 to 36 months of age at 40 New York City Department of Health well-baby clinics in various boroughs of New York City revealed that 21% of black children, 11% of Hispanic children, and 2% of white children had hemoglobin levels of 10.0 g/dl or less1 (Table 1). The subjects in this survey with hemoglobin levels below 10 g/dl also had low mean corpuscular hemoglobin concentration (MCHC) levels, low serum iron levels, and high total iron binding capacity levels. The percentages of all children with hemoglobin levels of 10.0 g/dl or less by age groups are shown in Figure 1. The peak incidence of iron deficiency anemia occurs at 10 to 15 months of age when it reaches an incidence of 30%. The incidence decreases with increasing age and drops to less than 5% at 36 months of age.

Numerous other studies have substantiated the high incidence of iron deficiency anemia in children under 36 months of age, both in American and non-American populations. Even cross-sectional studies involving middle class white populations indicate an incidence of anemia of between 1.4% and 6.3% (Table 2). Available figures indicate that the incidence has not been substantially altered in several decades.

It has been observed by many investigators that there is a higher prevalence of iron deficiency anemia in black children, as compared to white children Table I).4''0 Although no socioeconomic group is spared, the incidence of iron deficiency anemia in large population groups is inversely proportional to economic status.1 Both of the apparent ethnic and socioeconomic factors in the incidence of iron deficiency anemia are probably related to dietary habits, insofar as high iron content foods or iron-fortified baby foods are probably less available to black children in lower socioeconomic groups.

EVALUATION…

Iron deficiency anemia is the most common nutritional deficiency in children and is widespread in childhood populations throughout the world. It is especially prevalent in infancy. Many factors make infants and children vulnerable to develop negative iron balance and iron deficiency anemia. These include the fact that their diet often contains only trace amounts of iron, their limited ability to absorb dietary iron, their need of iron for growth, as well as the high prevalence of parasitism and gastrointestinal blood loss in some populations.

Iron, which is abundant in the earth's crust, has unique and subtle chemical properties and carries out a wide range of biological functions. It is critical for certain metabolic and enzymatic processes, is essential for growth, and plays a vital role in the structure of the hemoglobin molecule. It is responsible for the transport of oxygen (as hemoglobin and myoglobin), it participates in the activation of both molecular nitrogen and oxygen (as nitrogenases, oxygenases and oxidases) and plays a role in electron transport (as cytochromes). In addition, because of the propensity of iron to hydrolyse in aqueous solution, special molecules have been designed for its transport (transferrin) and storage (ferritin).

PREVALENCE OF IRON DEFICIENCY

Surveys of preschool children in many countries in the last half century have revealed widespread iron deficiency anemia. The incidence of the disorder has varied from population to population. Incidence is dependent upon a number of factors, such as the age group of the population selected, ethnic composition, dietetic habits, socioeconomic factors, presence of intestinal parasites, and the methods used for the detection of iron deficiency.

A number of surveys carried out in urban areas of the US over the last 20 years have revealed that the incidence of iron deficiency anemia in children between 6 and 36 months of age varies from 17% to 44%. A survey of 417 children from 6 to 36 months of age at 40 New York City Department of Health well-baby clinics in various boroughs of New York City revealed that 21% of black children, 11% of Hispanic children, and 2% of white children had hemoglobin levels of 10.0 g/dl or less1 (Table 1). The subjects in this survey with hemoglobin levels below 10 g/dl also had low mean corpuscular hemoglobin concentration (MCHC) levels, low serum iron levels, and high total iron binding capacity levels. The percentages of all children with hemoglobin levels of 10.0 g/dl or less by age groups are shown in Figure 1. The peak incidence of iron deficiency anemia occurs at 10 to 15 months of age when it reaches an incidence of 30%. The incidence decreases with increasing age and drops to less than 5% at 36 months of age.

Numerous other studies have substantiated the high incidence of iron deficiency anemia in children under 36 months of age, both in American and non-American populations. Even cross-sectional studies involving middle class white populations indicate an incidence of anemia of between 1.4% and 6.3% (Table 2). Available figures indicate that the incidence has not been substantially altered in several decades.

It has been observed by many investigators that there is a higher prevalence of iron deficiency anemia in black children, as compared to white children Table I).4''0 Although no socioeconomic group is spared, the incidence of iron deficiency anemia in large population groups is inversely proportional to economic status.1 Both of the apparent ethnic and socioeconomic factors in the incidence of iron deficiency anemia are probably related to dietary habits, insofar as high iron content foods or iron-fortified baby foods are probably less available to black children in lower socioeconomic groups.

EVALUATION OF PARAMETERS USED IN DIAGNOSIS OF IRON DEFICIENCY

Iron deficiency usually evolves slowly and progresses through several stages before it develops into frank anemia. The sensitivity of various iron measurements varies with the severity of iron lack and, on this basis, iron deficiency is conveniently divided into three stages. The earliest stage is storage mm depletion where iron reserves are lost but there is no decrease in iron supply to the developing red cell. The second stage is mm deficient erythropoiesis in which erythroid iron supply is diminished but the circulating hemoglobin is not significantly decreased. The third and final stage is overt iron deficiency anemia. The key laboratory measurements for each stage are listed in Table 3.

Table

TABLE 1HEMOGLOBIN LEVELS OF CHILDREN AGED 6-36 MONTHS ATTENDING WELL BABY CLINICS

TABLE 1

HEMOGLOBIN LEVELS OF CHILDREN AGED 6-36 MONTHS ATTENDING WELL BABY CLINICS

Although it is convenient to classify laboratory tests according to these three stages of iron deficiency, laboratory results often fail to conform to this pattern among individual patients. For example, certain patients may prove to have an iron-responsive anemia despite a normal serum ferritin or transferrin saturation. Such unexpected patterns of laboratory results may be confusing and at variance with the commonly considered concept of the development of iron deficiency anemia.

The accuracy of detecting iron deficiency anemia in population surveys can be substantially improved by employing a battery of laboratory measurements of the iron status and not relying exclusively on one parameter, such as hemoglobin level. Because of marked overlap of hemoglobin levels in anemia and normal populations, the definition of anemia based on hemoglobin levels alone results in a large number of false positive and false negative findings. Using the customary values to separate normality from iron deficiency, many potentially iron-responsive individuals may go undetected, while many of those identified as anemia may not prove to be iron deficient." Thus, large errors are introduced when the hemoglobin level is used as the only laboratory test. This overlapping of hemoglobin values is an obstacle to detecting those individuals whose hemoglobin is no more than 2 g/dl below their potential or normal level. There is less chance that hemoglobin values overlap into the normal range when anemia is more severe. For this reason, other laboratory tests of iron status used in conjunction with the hemoglobin level may be helpful in making the diagnosis.

The laboratory diagnosis of iron deficiency in infants and children requires special attention to the use of age-specific reference standards, since there are marked developmental changes in normal values not only for hemoglobin, but also for mean corpuscular volume (MCV), erythrocyte protoporphyrin, serum ferritin, and transferrin saturation. The derivation of reliable reference standards has been a particular problem for the pediatrie age group because subclinical iron deficiency anemia is common during early development, making it more difficult to select an appropriate reference population.

Figure 1. Percent of all children with a hemoglobin level of 10 g/dl or less by age group.

Figure 1. Percent of all children with a hemoglobin level of 10 g/dl or less by age group.

Hemoglobin Levels

Recently percentiïe curves for hemoglobin levels and MCV values derived from nonindigent white children living at sea level have been constructed (Figure 2).1Z This study was based on a total of 10,072 subjects. Of these, 7,489 children had determination of hemoglobin levels only, 1,358 had MCV values only, and in only 1,255 children was the serum ferritin, transferrin and MCV levels recorded. It is evident that the vast majority of these children had only hemoglobin levels and no other measurement of iron nutrition. Thus, it cannot be certain how "normal" this group was with reference to their iron nutrition, and how many subjects with alpha-thalassemia, β-thalassemia, or chronic infection were included in the normative data. Notwithstanding the potential fallacies, these data are the only percentiïe charts of hemoglobin and MCV available and as such are very useful in enabling us to assess various values in terms of normal percentiles.

The use of percentiïe curves avoids the errors inherent in tabulations, particularly near ages when there is an abrupt change in tabulated values. An approximate correction for altitude based on data from adults13 can be obtained by increasing the values by 4% for every 1,000 meters of elevation. These percentiïe charts are as useful as height, weight and head circumference percentiïe charts have proven to be in clinical practice. Future work has to be done to establish and refine the normalcy of the values that have been used to construct these charts.

An unresolved question relates to the application of the reference standards to blacks. Evidence to date indicates that blacks normally have a hemoglobin concentration that is about 0. 5 g/dl below that of whites at all ages, except perhaps in the perinatal period.1*·15 This difference does not appear to be due to iron deficiency, thalassemia trait, or differences in socioeconomic status. 15 If further studies confirm that this difference exists and that it is independent of health factors, it may be necessary to take this into account in applying reference standards to blacks.

Mean Corpuscular Volume

With reference to the MCV we have learned that accepting 80 fl (femtoliters or u.3) as minimum normal, based on adult values, is not correct and that the MCV is also subject to certain developmental and age-related factors. Red cells are normally larger at birth than in the adult, but red cell size decreases during the first 6 months of life. The red cell is smallest during the remainder of infancy and gradually increases in size during childhood. Mean values in girls are slightly higher than in boys after 7 years of age, but no sex differences are reported in adults. The lower limit for MCV is 70 fi between 10 and 17 months of age and there is a gradual increase of MCV with age; the lower limit for MCV is 74 fl between IYi and 4 years and 76 fl between 4 and 7 years (Figure 3). l6 In children with iron deficiency, the MCV is well below these normal limits for age, and the greater the certainty of the iron deficiency the lower the value for MCV (Figure 4).

Table

TABLE 2INCIDENCE OF IRON DEFICIENCY ANEMIA IN CROSS-SECTIONAL AND MIDDLE-CLASS POPULATIONS

TABLE 2

INCIDENCE OF IRON DEFICIENCY ANEMIA IN CROSS-SECTIONAL AND MIDDLE-CLASS POPULATIONS

Table

TABLE 3LABORATORY MEASUREMENTS OF IRON STATUS IN VARIOUS CONDITIONS LEADING TO IRON DEFICIENCY ANEMIA

TABLE 3

LABORATORY MEASUREMENTS OF IRON STATUS IN VARIOUS CONDITIONS LEADING TO IRON DEFICIENCY ANEMIA

As iron deficiency develops, the decrease in MCV appears to occur at about the same time as anemia and may actually show a greater relative deviation from normal. l6 A low MCV with anemia is most commonly associated with iron deficiency, but is also found in thalassemia minor. Iron deficiency anemia and thalassemia minor are often difficult to distinguish. The red cells in both conditions show moderate stippling and target and oval forms, but the larger number of hypochromic macrocytes in the blood smear and the familial hereditary pattern are important diagnostic features of thalassemia minor. However, the blood smears of patients with thalassemia minor and those with iron deficiency may both show a more or less uniform hypochromic microcytic picture, and other diagnostic aids are necessary.

Figure 2A. Percentile curves for hemoglobin concentration in infants and children. (Courtesy of Drs. RR. Dallman and M.A. Siimes, J Pediatr.12)

Figure 2A. Percentile curves for hemoglobin concentration in infants and children. (Courtesy of Drs. RR. Dallman and M.A. Siimes, J Pediatr.12)

Red cell indices provide a simple means of making a presumptive diagnosis without performing serum iron determinations or hemoglobin electrophoresis. The Mender /ormula, which consists of the MCV divided by the red cell count, gives one such index. Mentier formula values in excess of 13.5 strongly suggest that the patient has iron deficiency anemia, while values below 11.5 indicate thalassemia trait as the most likely diagnosis. 17 Another formula uses the discriminant /unction, expressed as: MCV - RBC - (5 X hemoglobin level) - 3.4. Rasitive values suggest a diagnosis of iron deficiency, while negative values indicate thalassemia trait as the cause of the microcytic anemia.18

Figure 2B. Percentile curves for mean corpuscular volume (MCV) in infants and children. (Courtesy of Drs. PR. Dallman and M.A. Siimes. J Pediatri12)

Figure 2B. Percentile curves for mean corpuscular volume (MCV) in infants and children. (Courtesy of Drs. PR. Dallman and M.A. Siimes. J Pediatri12)

These formulas are useful in the initial evaluation of patients. They should be used only as guides to whether a trial of iron or further hématologie investigations are indicated.

Blood Smear

Examination of a stained smear of blood is often said to be the simplest and most direct approach to diagnosing iron deficiency anemia. In one study, however, when hematologists examined blood smears from normal and iron deficient patients, iron deficiency was incorrectly diagnosed in 6% of normal smears, and was not recognized in 51% of smears in patients with proven iron deficiency anemia.19 Thus, while it is hetpfiil to examine the blood smear, a firm diagnosis of iron deficiency requires additional laboratory measurements, unless the iron deficiency is of marked degree.

Serum Iron and Iron-Binding Capacity

The time honored serum iron estimation as a measure of iron deficiency has serious limitations. The level of serum iron reflects the balance between several factors, including the iron absorbed, the iron utilized for hemoglobin synthesis, the iron released by red cell destruction, and the size of storage depots. At any particular time, the serum iron concentration represents a precise equilibrium between the iron entering and leaving the circulation.

The serum iron level is subject to circadian changes and it fluctuates by as much as 100 µ-g/dl during the day (Figure 5). Normal individuals have maximum serum iron levels in the morning with the lowest levels about 12 hours later, while the reverse is true in night workers. No variation has been found in children of ages 2 weeks to 20 months, whereas children from 3 to 14 years of age have a significant variation.20,21

Figure 3. Mean corpuscular volume changes with age. The shaded area represents plus or minus 2 standard deviation in adults. (Courtesy of Dr. M.A. Koerper, J Pedtatr.16)

Figure 3. Mean corpuscular volume changes with age. The shaded area represents plus or minus 2 standard deviation in adults. (Courtesy of Dr. M.A. Koerper, J Pedtatr.16)

Figure 4. Mean corpuscular volume in normal infants aged 10 to 17 months compared to infants in the same age range with iron deficiency or thalassemia trait. The shaded area represents the mean plus or minus 2 standard deviation in adults. (Courtesy of Dr. M.A. Koerper. J Pediatr.16)

Figure 4. Mean corpuscular volume in normal infants aged 10 to 17 months compared to infants in the same age range with iron deficiency or thalassemia trait. The shaded area represents the mean plus or minus 2 standard deviation in adults. (Courtesy of Dr. M.A. Koerper. J Pediatr.16)

It is of interest that the concentration of serum iron is higher in the morning and lower in the evening. This diurnal variation has been ascribed to the adrenal cortex and to variations in the autonomie nervous system.22 For this reason, blood for serum iron, should be drawn in the morning. 23 At 8 AM serum iron normally averages 140 pg/dl, yielding a saturation of 47%. This level falls throughout the day to a low of 40 µ??/dl by 10 PM or a saturation of 13%. A high degree of variability exists in adolescents with respect to serum iron and iron-binding levels. Boys tend to have higher serum iron levels than girls, slightly lower unsaturated ironbinding capacities, and a greater percentage saturation of the circulating transfetrins. Boys also tend to have higher mean hemoglobin levels, higher hematocrit values, and higher mean corpuscular hemoglobin concentration.24

Figuro 5. Circadian changes of serum iron. (Courtesy of Dr. L.D. Hamilton. Proc Sac Exp Biol Meo.20)

Figuro 5. Circadian changes of serum iron. (Courtesy of Dr. L.D. Hamilton. Proc Sac Exp Biol Meo.20)

Table

TABLE 4MEAN AND STANDARD ERROR OF SERUM IRON AND IRON SATURATION PERCENTAGE48

TABLE 4

MEAN AND STANDARD ERROR OF SERUM IRON AND IRON SATURATION PERCENTAGE48

Total serum iron is approximately 3 mg whereas the amount of iron which fluxes in and out of the plasma is in the range of 35 mg. Therefore, only slight change in the flux of iron from the reticulo-endothelial system results in big changes in the serum iron.

In addition, serum iron has a wide range of normal and varies significantly with age. Results of serum iron levels recently published are much lower than values hitherto accepted as being within the normal range. The means plus or minus the Standard Error (SEM) and 95% confidence limits at various ages for serum iron and iron saturation percentage are shown in Table 4.

Total iron-binding capacity (TIBC) may begin to increase as iron stores are depleted. An inverse correlation exists between serum ferritin and TlBC. While TIBC may reflect storage iron depletion, it is less sensitive to changes in iron stores than the serum ferritin.

The percent of transferrin saturation is a more sensitive index of iron status than serum iton alone, since total transferrin usually increases in iron deficiency whereas serum iron decreases. Furthermore, the percent saturation most accurately reflects the availability of iron for hematopoiesis. When transferrin drops below 15% or 16%, iron lack becomes limiting with reference to hemoglobin production. Despite this complication, transferrin saturation also has its limitations (Table 5) because it depends on plasma iron which is highly labile.

There is a wide variation of serum iron and transferrin saturation depending on age, sex, diurnal rhythms, dietary intake, laboratory methodology and other factors, resulting in a large normal range. For greater accuracy of diagnosis, serum iron and transferrin saturation should be used in conjunction with at least one other test of iion status. In addition, the developmental norms for serum iron and saturation, shown above, must be applied.

Free Erythrocyte Protoporphyrin

Free erythrocyte protoporphyrin (FEP) represents the penultimate stage in the biosynthetic pathway of heme, immediately prior to the incorporation of iron. Failure of iron supply will result in an accumulation of unutilized protoporphyrin in the normablast, which causes release into circulation of erythrocytes with high free protoporphyrin levels. Values obtained in one study of erythrocyte protoporphyrin (determined as jxg/dl of erythrocytes) were: hematologically normal subjects 15.5±8.3, subjects with latent iron deficiency 93 ±59.8, and patients with iron deficiency 159.2±96.5. The upper limit of normal free erythrocyte protoporphyrin is 40 µ-g/dl erythrocytes,25

The FEP/hemoglobin ratio is a useful index of iron deficiency. It increases exponentially in iron deficiency, accompanied by a decrease in both transferrin saturation and hemoglobin level. The FEP/hemoglobin ratio increases when the iron reserves are exhausted before anemia becomes apparent. In small children an elevation in the FEP/hemoglobin ratio is a better indicator of iron deficiency anemia than low transferrin saturation. 26 The FEP/hemoglobin ratio is normal in thalassemia trait and in renal anemia, but it may be elevated in sickle cell anemia and in chronic inflammatory processes.

An elevation of the FEP/hemoglobin ratio in blood reflects persistent iron deficiency in the bone marrow. Thus, the FEP/hemoglobin ratio remains elevated during iron therapy and returns to normal only after the majority of the red cells containing excess FEP formed during iron deficiency are replaced. Protoporphyrin is firmly bound to the hemoglobin in iron deficiency, just as it is in lead intoxication, and persists throughout the life span of the erythnxyte. Accordingly, the FEP/hemoglobin ratio is not subject to daily fluctuations and sudden changes such as occurs in transfemn saturation. Measurement of FEP and hemoglobin on filter paper provides a useful tool for the diagnosis of iron deficiency and, in populations at risk, for diagnosis of lead intoxication. Measurement of free erythrocyte protoporphyrin is carried out by a simple micromethod from fìnger puncture samples.27

The greatest elevation of free erythrocyte protoporphyrin levels is observed in lead intoxication, where values of up to 11 times normal may be observed. A level of FEP greater than 160 µg/dl of erythrocytes (equivalent to an FEP/hemoglobin ratio of 5.5 µg/g of hemoglobin) has been suggested as a convenient cut-off point for the detection of lead intoxication. In iron deficiency anemia the FEP/hemoglobin ratio is only moderately elevated and never exceeds 17.5 µg/g of hemoglobin (equivalent to 500 µg/dl of erythrocytes). An FEP/hemoglobin ratio in the range of 5. 5 µg to 17.5 µg per g of hemoglobin may be due either to iron deficiency anemia or to lead intoxication. Lead intoxication must be considered and should be ruled out by direct measurement of blood lead when prior exposure to lead is possible. When the FEP/ hemoglobin ratio is greater than 17.5 µg/g of hemoglobin, it indicates lead intoxication, with or without associated iron deficiency, and requires immediate medical attention. The only other syndrome associated with such high values of theFEP/hemoglobin ratio is the rare genetic disorder, erythropoietic protoporphyria, where severe cutaneous photosensitivity is present.

Table

TABLE 5LIMITATIONS OF TRANSFERRIN SATURATION DETERMINATION

TABLE 5

LIMITATIONS OF TRANSFERRIN SATURATION DETERMINATION

Table

TABLE 6INTERPRETATION OF FREE ERYTHROCYTE PROTOPORPHYRIN

TABLE 6

INTERPRETATION OF FREE ERYTHROCYTE PROTOPORPHYRIN

The FEP/hemoglobin ratio is not elevated in individuals with thalassemia trait;28 thus, measurements of FEP levels discriminate between the microcytosis of ß-thalassemia trait and that of iron deficiency anemia.

Free erythrocyte porphyrin is thus useful in distinguishing between the principal causes of microcytosis, namely, iron deficiency, alpha- or ß-thalassemia minor and lead poisoning. In both iron deficiency and in lead poisoning the level of free erythrocyte protoporphyrin is elevated, while it remains normal in thalassemia minor. It is much higher in lead poisoning than in iron deficiency. Table 6 lists the causes of raised levels of FEP and the advantages of FEP compared to transferrin saturation levels.

Plasma Ferritin

Recently, it has become possible, using a two-site immunoradiometric assay, to quantitate plasma ferritin concentrations. Ferritin is found mainly in the cytoplasm of reticuloendothelial cells and liver cells. It is normally thought to be an intracellular storage compound from which iron is mobilized into the transferrin-bound plasma pool. The level of serum ferritin reflects the level of body iron stores and is quantitative, reproducible, sensitive and requires only a small blood sample. Measurement of the serum ferritin level provides a non-invasive method for assessing a segment of the body ferritin pool. In normal subjects, 1 µg of ferritin per liter of serum represents about 8 mg of storage iron.29

Figure 6. Developmental changes in concentration of serum ferritin. Mean values (geometric means) in healthy populations are shown. At all ages, serum ferritin concentration below 10 or 12 µ-g per liter is indicative of depleted iron stores. (Courtesy of Dr. PR. Dallman, Am J Clin Nutr.47)

Figure 6. Developmental changes in concentration of serum ferritin. Mean values (geometric means) in healthy populations are shown. At all ages, serum ferritin concentration below 10 or 12 µ-g per liter is indicative of depleted iron stores. (Courtesy of Dr. PR. Dallman, Am J Clin Nutr.47)

During infancy and childhood, serum ferritin closely parallels the developmental changes in iron status. The mean cord blood concentration of plasma ferritin is 113 ng/ml. Beginning on the second day of life, there is a sharp increase in plasma ferritin to a mean level of 356 ng/ml at 1 month of age. High levels are maintained until 2 to 3 months of age, when iron stores are mobilized to meet the erythropoietic demands of an expanding hemoglobin mass and the level of plasma ferritin declines. Infants not receiving an available source of dietary iron are found to have no measurable ferritin level by 6 months of age, reflecting complete depletion of body iron stores. The mean value of plasma ferritin from 6 months through 15 years of age is 30 ng/ml (95% confidence limits: 7-142 ng/ml). This value is similar to the plasma ferritin concentration observed in iron-depleted adults. The low levels of plasma ferritin found in healthy children indicate that there is essentially no accumulation of storage iron during the period of childhood growth. In the adult, median concentrations of ferritin are 39 ng/ml in the female and 140 ng/ml in the male.30 Figure 6 shows the fluctuations in serum ferritin concentrations during development.

The mean concentration of serum ferritin is higher in men (123 ng/ml) than in women (56 ng/ml) with a range between 12 and 300 ng/ml.31 In patients with iron deficiency anemia, concentrations of serum ferritin are below 12 ng/ml and in patients with iron overload the concentration may be as high as 10,000 ng/ml.31

The serum ferritin concentration can be used both to evaluate iron stores in patients with suspected iron deficiency and to determine the level of stores after the completion of iron therapy.32 It is particularly useful for measuring iron stores in patients with chronic disease such as rheumatoid arthritis who may also have a hypochromic microcytic anemia.33 Thus, in patients with anemia due to chronic disease, the serum ferritin level is the most effective way to determine the need for iron therapy. Serum ferritin level has also been used in monitoring the iron status of patients with chronic renal failure on a regular dialysis regimen,34 and here too iron therapy can be controlled in a rational manner. Iron overload can be monitored by serial estimations of serum ferritin concentration; likewise, in patients with idiopathic hemochromatosis who are undergoing venesection the assay will indicate when body iron stores have been fully mobilized. In patients with thalassemia who have been on a regular transfusion regimen, the effect of chelation therapy can be assessed by the serum ferritin level.35

In iron deficiency, the serum ferritin assay seems to have an important advantage over serum iron and iron-binding capacity in that low values are almost invariably diagnostic. Iron deficiency anemia has sometimes been difficult to distinguish from the anemia of infection, since the serum iron and percent iron saturation can be low in both conditions.36 The fact that infection is associated with a normal or elevated serum ferritin concentration should add to its diagnostic value. It is uncertain whether a low serum ferritin concentration may precede a fall in the saturation of serum iron in children with inadequate dietary intake of iron, such as seems to occur in anemia due to blood loss.37

At all ages, a serum ferritin value below 10 or 12 µg/liter indicates depletion of iron reserve, but the overlap between normal and subnormal values is greater in women and in children at ages when ferritin values are normally low. A concentration of serum ferritin of less than 10 or 12 fig/liter is characteristic only of depleted iron stores. However, when inflammatory disease and iron deficiency coexist, serum ferritin values may be within the normal range. 38'40 One area of uncertainty is the reliability of the serum ferritin in diagnosing iron deficiency in infancy. Although a low serum ferritin indicates depleted iron stores,30,31,38,41 a normal serum ferritin is often present in cases of iron-responsive anemia.

Bone Marrow Iron

Bone marrow iron estimation is a time-honored method for assessing storage iron. The presence of bone marrow iron excludes iron deficiency. This method is only utilized when difficulty arises in the etiology of anemia and bone marrow examination is indicated as part of the hematologic evaluation. Disparity exists between aspirate specimens and biopsy specimens, probably as a result of assessing marrow aspirates containing insufficient marrow stroma. The absence of marrow iron on an aspirated specimen is not unequivocal evidence of absent stores. When evaluating bone marrow iron it is important to examine both unstained and stained aspirates and to obtain generous amounts of stroma, and to confirm the findings on marrow biopsy when available. This method for the diagnosis of iron deficiency anemia is used only when considerable difficulty exists in the elucidation of the cause of anemia in a patient.

Therapeutic Trial

The most reliable criterion of iron deficiency anemia is the hemoglobin response to an adequate therapeutic trial of iron. A therapeutic trial allows the recognition of individuals whose hemoglobin values, although within the reference range, are low for those individuals.

Under ordinary clinical circumstances, the concentration of hemoglobin should be about two-thirds corrected toward the child's potential value after a 1-month trial of iron therapy. If anemia persists despite good compliance, other causes of anemia should be considered.

Table

TABLE 7DIAGNOSTIC TESTS FOR IRON DEFICIENCY ANEMIA

TABLE 7

DIAGNOSTIC TESTS FOR IRON DEFICIENCY ANEMIA

An increase of 1 g/dl is often considered significant, but a 2 g/dl increase in hemoglobin level is much more reliable. The therapeutic trial has been helpful in detecting subclinical iron deficiency in prevalence studies, and in determining the relative importance of iron lack when the cause of anemia is multifactorial. In a clinical setting, the value of a therapeutic trial depends on the type of patient. If anemia is detected in the office or clinic in a patient at high risk of developing iron deficiency, a practical inexpensive approach is to place the patient on oral iron and to determine the hemoglobin response in 3 to 4 weeks. The therapeutic trial is not helpful in hospitalized patients because of the time required to make a diagnosis, but it can be advantageous in clinical investiga' tion. In a patient with several possible causes of anemia, such as in rheumatoid arthritis or chronic renal failure, a therapeutic iron trial provides concrete evidence of clinically significant iron deficiency anemia.

In the final analysis, the response to iron therapy is the proof of correctness of the diagnosis of iron deficiency. The physician may not have access to all diagnostic techniques, or on clinical grounds the probability of iron deficiency anemia is so high that the patient's response to therapy may become of primary diagnostic importance. Iron administration in such a therapeutic trial should be by the oral route only and response should be followed very carefully. A reticulocytosis with a peak observed between the 7th and 10th days should occur and a significant rise in hemoglobin should be seen. The absence of these changes must be taken as evidence that iron deficiency is not the cause of the anemia. In the absence of a rise of the hemoglobin level and reticulocytosis, iron therapy should be discontinued and further diagnostic studies implemented. Table 7 summarizes the diagnostic tests available in the investigation of iron deficiency anemia.

Table

TABLE 8DISORDERS ASSOCIATED WITH HYPOCHROMIA

TABLE 8

DISORDERS ASSOCIATED WITH HYPOCHROMIA

Differential Diagnosis of Iron Deficiency

The presence of anemia in a child in whom there is absence of other hematologic abnormalities is most likely due to iron deficiency. The presence of microcytosis (with a mean corpuscular volume as determined on the Coulter model S of less than 70 fl) and hypochromia (with a mean corpuscular hemoglobin concentration less than 30%) is consistent with the diagnosis of iron deficiency anemia. The easiest and most reliable diagnostic criterion is the response of the anemia to iron, as in a therapeutic trial. If the iron is taken and the hemoglobin level has not increased within 3 weeks time, the diagnosis of iron deficiency anemia is probably erroneous unless bleeding has occurred.

Although hypochromic anemia in children is usually due to iron deficiency, it is not necessarily due to this cause. A list of the causes of hypochromia are given in Table 8. In some of these cases there is an inability to synthesize hemoglobin normally in spite of a plentiful supply of iron. In unusual or obscure cases of hypochromic anemia, it is necessary to do additional investigations such as determination of serum iron level, total iron binding capacity, serum ferritin, hemoglobin electrophoresis, and examination of the bone marrow for stained iron in order to establish the cause of the hypochromia. Table 9 lists the investigations employed in the differential diagnosis of hypochromia and Figure 7 depicts a flowsheet for the diagnosis of hypochromic anemia.

Table

TABLE 9INVESTIGATIONS IN THE DIFFERENTIAL DIAGNOSIS OF HYPOCHROMIC ANEMIA

TABLE 9

INVESTIGATIONS IN THE DIFFERENTIAL DIAGNOSIS OF HYPOCHROMIC ANEMIA

Figure 7. Flow sheet for the diagnosis of hypochromrc anemia.

Figure 7. Flow sheet for the diagnosis of hypochromrc anemia.

Confusion of iron deficiency anemia with homozygous thalassemia major may occur only in the infant in the first year of life. The appearance of large, pale, extremely thin erythrocytes with irregularly distributed hemoglobin and scattered normoblasts interspersed among microcytes in the blood smear is in contrast with the more uniform appearance of microcytes of iron deficiency anemia. The significant splenomegaly and the presence of the thalassemia trait in both parents and siblings further differentiates thalassemia major from iron deficiency anemia.

In addition to making a diagnosis of iron deficiency anemia, it is incumbent on the physician to demonstrate its cause. The history should take into account all factors related to the development of iron deficiency. This should include search for conditions resulting in low iron stores at birth, a careful dietary history, and consideration of all factors leading to blood loss. The commonest site of bleeding is into the bowel and the most important investigation is examination of the stool for occult blood. If occult blood is found, its cause should be established by examination of stools for ova, rectal examination, sigmoidoscopy, barium enema, upper gastrointestinal series, and technetium-99 pertechnetate scan for detection of a Meckel's diverticulum. Occasionally, gastroscopy and colonoscopy are required.

Negative guaiac tests occur particularly if bleeding is intermittent and for this reason occult bleeding should be tested for on at least five occasions when gastrointestinal bleeding is suspected. The guaiac test is only sensitive enough to pick up more than 5 ml of occult blood. In menstruating females excessive uterine bleeding, epistaxis, renal blood loss (hematuria), and on rare occasions bleeding into the lung (idiopathic pulmonary hemosiderosis and Goodpasture's syndrome) may all be causes of iron deficiency anemia. Bleeding into these areas requires specific investigations designed to detect the bleeding and to determine its cause.

SUMMARY AND CONCLUSION

Iron deficiency is the most common nutritional deficiency in children and is widespread in childhood populations throughout the world. Although many sophisticated tests have been devised for the diagnosis of iron deficiency the most reliable criterion of iron deficiency anemia is the hemoglobin response to an adequate therapeutic trial of iron. Following the reticulocytosis peak hemoglobin rises at an average of 0. 25 to 0. 4 g/dl/day and hematocrit at a rate of 1% per day. If the response to iron falls short of this response other causes of the anemia should be sought by detailed hematologic investigation. In addition to making a diagnosis of iron deficiency anemia it is incumbent on the physician to demonstrate its cause.

REFERENCES

1. Lanzkowsky P: Iron deficiency anemia. Pediatr Ann 1974; 3:6.

2. Fuerth JH: Incidence of anemia in full-term infants seen in private practice. J Pediatr 1971; 79:562.

3. Owen GM, Nelson CE, Carry PJ: Nutritional status of preschool children; hemoglobin, hematocrit and plasma iron values. J Pediatr 1970; 76:761.

4. Danneckcr D: Anemia in selected Allegheny Counry child health conference populations. Allegheny County Health Department. Pittsburgh, Pennsylvania.

5. Davis LR, Maren RH, Sarkany 1; Iron deficiency anemia in European and West Indian infanti in London. Br Med J 1960; 2:1426.

6. Hillman RW: Relationship of race and sex to the frequency of local tissue changes suggestive of malnutrition; the five year experience of a district health center nutrition clinic in New York City. Am J CIm Nutr 1962; 10:410.

7. Lanzltowsky P: Iron Deficiency: A Public Healthi Problem. Evansville, Indiana, Mead Johnson Company, 1975.

8. Munday B, Shepherd ML, Emerson L, et. al: Hemoglobin differences in healthy white and Negro infants. Am J Dis Child 1938; 55:776.

9. Hist RT, Bhardwaj B, Lamluwsky P: Prevalence and pathogenetic factors of anemia in children in New York City. Paper presented to American Pediatrie Ambulatory Care Society, Atlantic City, New Jersey. April 1968.

10. Vitale JJ, Restrepo A, Valez H, et al: Secondary folate deficiency induced in the rat by dietary iron deficiency. J Nutr 1966; 88;315.

11. Garby L, Imeli L, Werner I: Iron deficiency in women of fertile age in a Swedith community, Ill. Estimation of prevalence based on response to iron supplementation. Acta Med Scand 1969; 185:113.

12. Dallman PR, Slimes MA: Percentile curves for hemoglobin and red cell volume in infancy and childhood. J Pediatr 1979; 94:26.

13. Cartwright GEi Diagnostic Laboratory Hemaiology (ed 4). New Yotk. urline and Stratton. 1968.

14. Garn SM. Smith N]. Clark DC: Litelong differences in hemoglobin levels between blacks and whites. J Noil Med Assoc 1975; 67:91.

15. Dallman PR, Barr GD. Alien CM, et al: Hemoglobin concentration in white, black and Oriental children; is there a need for separate criteria in screening for anemia? Am Clin Nutr 1978;31:377.

16. Koerper MA, Menner WC. Brecher G, et al· Developmental change in red blood cell volume; implications in screening infants and children for iron deficiency and thalassemia trait. J Pediarr 1976; 89:580-583.

17. Mentier WC Jr. Differentiation of iron deficiency from thalassemia trait. Lancet 1973; 1:882.

18. England )M, Fraser P: Differentiation of iron deficiency from the thalassemia trait by routine blood-count. Lancet 1973; 1:449.

19. Fairbanks W: Is the peripheral blood film reliable for the diagnosis of iron deficiency anemia. Am J Clin Rufwl 1971; 55:447-451.

20. Hamilton LD, Gubler CJ, et al: Diurnal variation in plasma iron level in Man. Ptoc Soc Exp Biol Med 1980; 75:65-68.

21. SchwartiE, BaehnerRL: Diurnal variation of serum iron in infants and children. Acta Pediatr Scan J 1968; 57:433-435.

22. Bowie JW, Tauxe WN, Sjoberg WE Jr, et al: Daily variation in the concentration of iron in serum. Am J Clin Pathol 1963; 40:491.

23. Dallman PR: The nutritional anemias, in Nathan DG, Oski FA (eds). Hematulogy oi infancy and Childhrood, Philadelphia, W.B. Saunders Company. 1971.

24. Seltzer CC, Wenzel BJ, Mayer J: Serum iron and iron-binding capacity in adolescents. I. Standard values. Am J Clin Nutr 1963; 13:343.

25. Dagg JH, Goldbert A, Lockhead A: Value of erythrocyte protoporphyrin in the diagnosis of latent iron deficiency. Br J Haematol 1966; 12:326.

26. PiomelliS, Brickman A, Carlos E: Rapid diagnosis of iron deficiency by measurement of free crythrocyte prophynra and hemoglobins; the FEB/hemoglobin ratio. Pediatrics 1976; 57:136.

27. Piomelli S, Davidow B: Free erythrocyte protoporphyiin concentration; a promising screening test for lead poisoning. Pediatr Res 1972;6:366.

28. Stockman JA, Weiner LB. Smart MJ, et al: The micrameasurement of free erythrocyte ptophyrin (FEP) as a means of screening for beta-thalassemia minor in subjects with microcvtosis. Blood 1973; 42:990.

29. Waken GCi Miller F, Worwood M: Serum ferritjn concentration and iron stores in normal subjects. J Clin Pathol 1973; 26;770.

30. Sümes MA, Addiego JE, Dallman PR: Ferritin in serum: Diagnosis of iron deficiency and iron overload in infants and children. Blood 1974; 43:581.

31. Jacobs A, Worwood M: Ferritin in serum; clinical and biochemical implications. N End J Med 1975; 292:951.

32. Bentley DP, Jacobs A: Accumulation of storage iron in patients treated fot irondeficiency anaemia. Br Med 1975; 1:64.

33. Bentley DP, Williams P: Serum ferriti n concentration as an index of storage iron in rheumatoid arthritis. J Clin Pathol 1974; 27:786.

34. Hussein S, Prieto J, O'Shea M, et al: Serum ferritin assay and iron status in chronic renal failure and haemodialysis. Br Med J 1975; 1:546.

35. Letsky EA, Miller F, Worwood M, et al: Serum ferritin in children with thalassemia regularly transfused. J Clin Pcnhol 1974; 27:652.

36. Bainton DF, Finch CA: The diagnosis of iron-deficiencv anemia. Am J Med 1964; 37:62.

37. Jacobs A, Miller F. Worwood M. et al: Ferritin in the serum of normal subjects and patients with iron deficiency and iron overload. Br Med J 1972; 4:206.

38. Lipschitz DA, Cook JD, Finch CA: A clinical evaluation of serum ferritin as an index of iron stores. N Engl J Med 1974; 290:1213.

39. Elin RJ. Wolff SM, Finch CA: Effect of induced fever on serum iron and ferritin concentrations in man. Blood 1977; 49:147.

40. Koerper MA, Stempel DA, Dallman PR: Anemia in patients with juvenile rheumatoid arthritis. J Pediatr 1978; 92:930.

41. Cook JD. UpschitzDA, Miles LEM, et al; Serum ferritin as a measure of iron stores in normal subjects. Am J Clin Nutr 1974; 27:681.

42. Serbie J, Olatunbosum D, Corbett WEN, et al: Cobalt excretion test for the assessment of body iron stores. Can Med Assoc J 1971; 104:777.

43. Thomson ABR, Shaver C, Lee DJ, et al: Effect of varying iron stores on site of intestinal absorption of cobalt and iron. Am J Physiol 1971; 220:674.

44. Albert LS, Ludwig J, Olatunbosum D: Alteration in cobalt absorption in patients with disorders of iron metabolism. Gastroenterology; 1969; 56:241.

45. Balcenak SP, et al: Measurement of iron stores using deferoxamine. Ann Intern Med 1968; 68:518.

46. Hallberg L, Hedenberg I, Weinneld A: Liver iron and desterrioxamine-induced urinary iron excretion. Scandd J Haematol 1966; 3:85.

47. Dallman PR, Siimes MA, Stekel A: Serum ferritin levels in healthy children. Am J Clin Nutr 1980; 33:86.

48. Koerper MA, Dallman PR: Serum iron concentration and transrerrin saturation are lower in normal children than in adults, Pediatr Res 1977: 11:473.

TABLE 1

HEMOGLOBIN LEVELS OF CHILDREN AGED 6-36 MONTHS ATTENDING WELL BABY CLINICS

TABLE 2

INCIDENCE OF IRON DEFICIENCY ANEMIA IN CROSS-SECTIONAL AND MIDDLE-CLASS POPULATIONS

TABLE 3

LABORATORY MEASUREMENTS OF IRON STATUS IN VARIOUS CONDITIONS LEADING TO IRON DEFICIENCY ANEMIA

TABLE 4

MEAN AND STANDARD ERROR OF SERUM IRON AND IRON SATURATION PERCENTAGE48

TABLE 5

LIMITATIONS OF TRANSFERRIN SATURATION DETERMINATION

TABLE 6

INTERPRETATION OF FREE ERYTHROCYTE PROTOPORPHYRIN

TABLE 7

DIAGNOSTIC TESTS FOR IRON DEFICIENCY ANEMIA

TABLE 8

DISORDERS ASSOCIATED WITH HYPOCHROMIA

TABLE 9

INVESTIGATIONS IN THE DIFFERENTIAL DIAGNOSIS OF HYPOCHROMIC ANEMIA

10.3928/0090-4481-19850901-07

Sign up to receive

Journal E-contents