How well can iron status and health be measured and when should they be evaluated?
A variety of tools have been developed for evaluating the iron nutritional status and metabolism of women of reproductive age (WRA) prior to and during pregnancy, and for evaluating the likelihood of a pregnant woman producing a baby with desirable iron conditions. Specifically, the biochemical indicators of the iron status of non-pregnant WRA are unique in that they can measure with accuracy the different metabolic stages of this metal, from its absorption, transport once absorbed, utilization in heme formation, storage, reutilization, and its deficit or excess at the cellular level.11–13 Important and simple surrogates indicating a prolonged deficit of iron are the measurement of hemoglobin (Hb) concentration and the characteristics of red blood cells. Unfortunately, they lack the desired specificity. Specialized biochemical techniques have been developed to measure iron stores and factors that control iron absorption and the regulation of iron metabolism.14 Also, body iron reserves can be measured by the superconducting quantum interference device (Ferritometer®), of which there are only four worldwide (one is at Children's Hospital Oakland Research Institute). This non-invasive device effectively measures liver iron content, particularly when it is elevated. Unfortunately, its use is expensive, body characteristics may interfere with its performance, and its precision decreases in the lower ranges of liver iron stores when compared to liver biopsy iron measurements.15 In any case, there are no studies on measurements of iron stores in pregnant women using this non-invasive method.
How well do the commonly available biochemical techniques function in defining mild-to-moderate iron deficiency (ID) in pregnancy? Unfortunately, not that well. There is overwhelming evidence indicating that the great majority of cases of gestational anemia from the second trimester onwards are due to ID because of the overall higher demands for iron during the second and third trimesters. However, the definition of gestational anemia and of hemoconcentration, as well as that of ID and excess, are not that well defined, particularly considering that ID of nutritional origin and anemia associated with it are generally mild to moderate.16,17 Severe anemia is generally the result of complicating factors.
An important example of the relationship between a woman's iron status at or near the time of conception and the occurrence of gestational anemia later in pregnancy is a publication from Mexico by Kaufer and Casanueva.18 Their report showed very clearly that women with absent or with low iron reserves (serum ferritin < 20 µg/L) at the beginning of pregnancy developed gestational anemia later in pregnancy more than five times more frequently than women with larger iron reserves at gestational week 10. Similarly, other studies19 have reinforced the importance of pre-pregnancy iron nutrition for the prevention of gestational ID and anemia during the course of pregnancy and for the welfare of offspring. In anemic pregnant women, ID is generally the primary cause but this may not be the case in women with concomitant malaria, HIV, and other infections and in the presence of hemoglobinopathies. Importantly, about one-third of non-pregnant WRA worldwide are iron deficient, and even in industrial societies close to half have unsatisfactory iron reserves.20,21
The specificity with which iron status is measured during pregnancy is limited because of several changes that affect the indicators of iron nutrition. For example, Table 3 presents the evolution of the mean values of indicators commonly used during pregnancy, based on the classic longitudinal studies by Milman et al.,22 Romslo et al.,23 Svanberg,24 Sjostedt et al.,25 Puolakka,26 and Scanlon et al.,27 with female participants who were healthy, non-anemic, and iron-sufficient. In these studies, women received the following daily doses of iron supplements from early pregnancy: 20–80 mg,22 100 mg,23 and 200 mg24, 50–200 mg,26 or placebo23,24. Despite the various levels of iron supplementation, in the Milman sample,22 there were no differences in the parameters measured except for serum ferritin levels, that decreased continuously up to gestational week 32 among women receiving 20 mg of iron daily, while serum ferritin levels showed an increasing trend at that gestational week among women receiving more supplementary iron. Therefore, all the values in that publication are pooled as a single supplemented group. The geometric mean values for Hb do not differ from those reported by Scanlon27 in a cross-sectional study of iron-supplemented women in the United States and from the median values considered normal by the Centers for Disease Control (CDC),28 as based on four studies with a small number of healthy supplemented women in northern Europe. The CDC values for Hb concentration at week 24 are 115 g/L. These values are lower than 118 and 117 g/L reported by Scanlon27 and Milman22 respectively. The indicators of iron status in the women receiving supplements and in those receiving placebos were all modified in the direction of having lower iron status up to gestational weeks 30–32; serum ferritin and serum iron both declined (serum iron less dramatically in the iron-supplemented groups), while total iron-binding capacity rose more in the placebo group, making the percent saturation of total iron-binding capacity lower. Free erythrocyte protoporphyrins rose moderately at the same gestational weeks. Hb concentrations also dropped moderately in both groups until about gestational weeks 30–32, when those in the supplemented groups rose while those in the placebo groups continued to fall. Most other indicators tended to improve after weeks 32–34 in the supplemented group only.
Table 3. Evolution of the mean blood values of selected iron nutrition indicators during pregnancy in normal women receiving antenatal iron supplements.a
|Parameter||Normal values for females||Males|
|1st T.*||2nd T*||3rdT*|
|Serum iron (µg/dL)||40–150||106||90||80||50–175|
|Saturation of TIBC (%)||20–50||30.5||20.9||15.9||20–50|
|FEPµg/100 mL RBC||15.71 ± 7.26||15.7||18.2||23.2||15.71 ± 7.26|
|FEP/Hb ratio||0.44 ± 0.21||0.52||0.61||0.78||0.44 ± 0.21|
|Serum ferritin µg/L||12–150||32||18||21||12–300|
|Soluble Transferrin Receptors (mg/L) (STfR)||5.36 ± 0.82||5.4||8.0||8.6||5.36 ± 0.82|
|Red cell diameter width coefficient of variation||11–15||11–15||11–15|
|Reticulocyte Hemoglobin contents (pg)||26–32||>26 pg/L||26–32|
|Bone marrow iron**|
|Gestational week|| ||12 ± 1||35 ± 1|| |
|% in Placebo group respectively**||0, +, ++, +++ or ++++||2, 10, 31,53, 4**||79, 21, 0, 0, 0**|| |
|% in Supplemented group respectively**||0, +, ++, +++++++||2, 2, 38, 44, 14**||41, 24, 35, 0, 0**|| |
What is important is that, even when “normal” European women were receiving as much as 200 mg of iron daily, the biochemical indicators pointed in the direction of poorer iron status. Similarly, the marked decline in bone-marrow iron (Table 3) even in women receiving 200 mg of iron daily24 by gestational week 35, strongly suggests there are modifications in iron metabolism and in several of the indicators of iron nutritional status during pregnancy that need clarification. Notably, even though these indicators suggest an iron-deficient state, particularly late in gestation (week 36), the incorporation of absorbed iron into Hb was lower than in the iron-deficient, non-pregnant women,24 which suggests effective competition for iron utilization from the feto-placental unit.
The cutoff points for Hb levels in the diagnosis of gestational anemia, as proposed by the CDC and the World Health Organization (WHO),28,29 do not coincide with the Hb levels reported as predictors of pregnancy outcome. Specifically, several studies indicate that the outcome of pregnancy is best when maternal Hb levels during the second and third trimesters are higher than 90–100 g/L and lower than 125 g/L.30–36 The lower concentrations are below the CDC/WHO cutoff levels for anemia (110 g/L in the first and third trimesters and 105 g/L during the second trimester).
Whether or not anemia is a cause of low birth weight or premature delivery and of perinatal risk is controversial, depending on if it is due to ID in the first trimester of pregnancy, as described in the Camden study,37 or later, or if it is mild, moderate, or severe. Mild-to-moderate maternal anemia (Hb > 90 ≤ 100 g/L) has little, if any, effect on raising the incidence of premature delivery or low birth weight, including small-for-gestational-age (SGA) deliveries.38–40 However, the more severe the anemia, the higher the obstetric and perinatal risks.38 On the other hand, the risk for low birth weight and premature delivery begins to increase steeply when Hb levels rise above 125 g/L in the second and third trimesters.36
It is important to keep in mind that in the case of black women, more cases of anemia are observed if the distributions and cutoff points derived from the northern European and CDC population studies are used because Hb levels in healthy black women are 0.8 g/L lower.29 This means ethnicity must be taken into account when Hb level is a variable being considered.
The indications suggestive of iron depletion during gestation, even when supplementation with high iron doses is administered, was corroborated by the studies of Svanberg et al.,24 who demonstrated a marked decline in bone-marrow iron among women supplemented with 200 mg of iron daily, as indicated above. Moreover, his studies clearly show increased iron absorption in the third trimester of pregnancy. In contrast, in bone marrow samples taken 9 weeks post partum, the supplemented women had higher levels of iron in bone marrow than was present early in pregnancy, indicating that the high levels of supplemented iron were partially accumulated despite the blood biochemical indicators reductions in bone marrow iron, suggesting progressive depletion of the metal during pregnancy (Figure 1).
Figure 1. Bone marrow iron during and after pregnancy.* Arbitrary scale by the total sum of points, assigning values of 0, 1, 2, 3, and 4 points to percent cases with 0, +, ++, +++, and ++++ iron staining in bone marrow at each occasion. Prepared from Svanberg's data.24
Download figure to PowerPoint
The well-known changes in plasma and red-cell volumes do not fully explain the changes in Hb concentration and in the biochemical indicators of iron status compared to their behavior in non-pregnant women, except that according to Hytten6 iron-supplemented pregnant women increase their total red cell volume by 400–450 mL compared to non-supplemented women, who increase their total red cell volume by about 250 mL. The post-partum resorption of the larger Hb mass may explain the post-partum increase in bone marrow iron among the supplemented women.
The parallel changes in bone marrow iron observed in the groups in Figure 1 are quite striking. However, the components of total points in each are different in that 79% of the non-supplemented group at week 35 had no bone marrow iron while 41% of the supplemented group had no bone marrow iron; in the non-supplemented group in the post-partum period, only 9% lacked bone marrow iron and 47% had “3 plus 4 plus” iron estimates (3 and 4 being +++ and ++++ iron, respectively), while in the supplemented group, the same bone marrow iron categories were 0% and 92% respectively.
A new form of estimating body iron, first proposed by Skikne et al.41 and Cook et al.,42 consists of the logarithm of the ratio of soluble transferrin receptors to serum ferritin. This has facilitated the estimation of iron stores as compared to previous methods. The two groups of researchers demonstrated that the logarithm of this ratio was linearly correlated with body iron stores and their deficit in mg/kg. This method is extremely useful and sufficiently accurate and precise, except at the extremes of iron excess and deficiency. Moreover, the plotting of body iron (mg/kg body weight) scores estimated using this method, versus the percent cumulative frequency distribution of populations, reveals deviation of the linear regression to indicate a higher prevalence of ID at the high values of this ratio in a population of US women between the ages of 20 and 45 years; this indicates the presence of ID in this female population while the straightness of the regression line is conserved in the case of a US male population, as is to be expected due to the absence of ID in this group. Similar results were obtained using a previous algorithm to estimate iron status.43,44
When this method is applied to pregnant women, it appears that 50% of pregnant Jamaican women are iron deficient, yet the cumulative frequency line is completely straight without showing any deviation in the deficient status compared to the sufficient status. This suggests that iron depletion does not exist beyond a normal depletion determined by this method, which displays a normally distributed continuum rather than a tail that deviates from the straight line indicating a true state of iron deficiency. This may be due to the changes in indicators that take place during pregnancy, whereby serum ferritin levels decline even when women are ingesting large doses of iron supplements, and transferrin receptor levels increase as a consequence of increased erythropoiesis45,46; this affects the ratio independently of true iron status. New technologies also allow study of the regulation of iron metabolism, beginning with iron absorption, by measuring regulatory proteins like hepcidin47,48 and iron availability for erythropoiesis by the characteristics of red cells and of reticulocytes46,49 as well as by estimating the hypoxic state and the response to that stimulus by measuring erythropoietin.50 Hypoxic inducible factor (HIFα2) and gestational-associated hormonal changes also modify erythropoietic activity.51–53
Maternal iron status after delivery and during lactation depends on the iron status during gestation, with the additional consequences of blood loss during delivery and of the reutilization after childbirth of the iron incorporated into the mother's increased red cell mass. The indicators of iron status readjust a few weeks after delivery, returning to their original pre-pregnancy meaning and the estimate of body iron can again be measured with confidence.
In essence, researchers are now uniquely able to quantify the iron content of most non-pregnant persons thanks to the development of technology that, when properly used, can determine not only a deficient or an excessive level of this metal but also how the level evolves over the course of life events and diverse interventions. Unfortunately, in the case of the pregnant woman, technological capabilities are still limited, with obstacles to determining the relationship between maternal iron status and fetal-newborn iron status remaining, as explained below. If possible, studies using the superconducting quantum interference device in pregnant women could be very revealing if different obstacles posed by maternal and fetal anatomical/structural changes can be overcome.
The estimation of iron reserves of the newborn is not that simple either. It is based on total Hb-iron, serum ferritin concentration, body weight, and functional body iron content estimated by a constant per kg body weight.9,10