81
Clinica Chimico Acta, 83 (1978) @ Elsevier/North-Holland
81-91
Biomedical
Press
CCA 9048
SERUM FERRITIN
H.G. VAN
AND IRON STORES DURING PREGNANCY
EIJK ‘+*, M.J. KROOS
*, G.A. HOOGENDOORN
b and H.C.S. WALLENBURG
b
a Department of Chemical Pathology and b Department of Gynaecology and Obstetrics, Medical Faculty, Erasmus University Rotterdam, Rotteniam (The Netherlands) (Received
July 15th,
1977)
summary The iron status of two groups of pregnant women was investigated. One group did not receive iron (group B), the other received 100 mg iron/day (group A). 1. In all individuals concentrations of hemoglobin, serum iron, transferrin and serum ferritin were determined at regular intervals from the third month until delivery and at 3 months after delivery. The same determinations were performed in cord blood. 2. Changes in iron status appeared to be less in individuals with iron supplement than in those without iron supplement. A fall in Hb, serum iron and serum ferritin is observed in all individuals. 3. Three months after delivery the Hb concentration has generally returned to the normal female value, but the serum ferritin concentration is still very low. 4. The fetus does not discriminate as to the iron status of the mother. In both groups (A and B) cord blood values appeared to be not significantly different.
Introduction Ferritin is the main storage compound for iron in the human body and is present mainly intracellularly in the reticuloendothelial system of the liver, the spleen and the bone marrow. Intracellular ferritin can be identified in histologic sections of bone marrow smears by staining with Prussian blue. A small amount is normally found in circulating plasma in concentration varying between 10 and 200 pg/l. In men the serum concentration is twice that of women, suggesting that the serum level reflects total body stores (liver, spleen, bone marrow, *
To whom
correspondence
should be addxeued.
82
Hb). In case of iron deficiency or iron overload the serum ferritin concentration is related to the concentration in the tissues. It should be noted, however, that in patients with tissue damage (acute and chronic liver disease) or malignancy (leukemia) the relationship with the iron status may be disturbed by variations in abnormal production or release of ferritin. During pregnancy an appreciable amount of iron is transferred from the mother to the fetus. At term, the human fetus contains 200-400 mg iron. The mechanism by which iron is transferred from the mother to the fetus across the placenta is poorly understood, notwithstanding a number of investigations in various animals [ 11. It is likely that the mobile iron pool in the liver is involved in the mobilization of iron. The object of the present study was to evaluate the significance of the determination of serum ferritin as a measure of the iron reserves during pregnancy and, therefore, as an indication for the application of iron therapy. A practical advantage of the radioimmunoassay of serum ferritin is that it provides a virtually noninvasive method for measuring a segment of the body ferritin pool [ 2,3]. Materials and methods Thirty pregnant women, attending the University Hospital/Obstetrical Clinic, were randomly selected for this study. The course of pregnancy and delivery was uncomplicated in all cases. Fifteen women (group A) received 100 mg iron daily (as FeSOj,) from the third month of pregnancy until delivery. Fifteen women did not receive any iron supplement (group B). Venous blood samples were obtained between 14.00 and 16.00 hour every three or four weeks from the third month until delivery and at three months after delivery. Five ml samples were collected in polystyrene tubes containing Sequestrene@ as an anticoagulant. Mixed cord blood was obtained immediately after delivery. The hemoglobin concentration was determined with a Coulter Counter Model 5, The inst~ment was calibrated with hematology reference control 4C (Coulter Diagnostics Inc., Hialek, Fla.). Serum iron and transferrin concentrations were determined as described previously [ 41. For the determination of serum ferritin a specific highly-sensitive solid phase radioimmunoassay was used, adopted from Ha&day [ 51, as reported earlier in detail [ 61. The reproducibility of all methods used was more than 95%. Identical variables from the groups A and B were statistically compared using the Wilcoxon test (two-tailed). Results We investigated the Hb, Tr, serum iron and serum ferritin concentrations in maternal blood at term, in maternal blood twelve weeks after term (supposing that the Hb values in this blood would be equal to those in the blood at time zero, before pregnancy), in cord blood at term and a time-course of I-Ib, Tr, serum iron and serum ferritin in about 15 pregnent women with (group A) and 15 women without (group 3) any iron supplement during their six-month follow-up period until delivery.
83
A mmot/l
Hemoglobin
M*
104
9.0
7’0
I 40
42
40
42
Weeks
B
mmol/l
Hemoglobin
10
16
M-
22
2a
34
Fig. 1. A time course of hemoglobin. M+. maternal blood with iron therapy; M-, iron therapy; -, mean value for each group (15 persons).
#oks
maternal blood without
84
(A Serum Fe
M+
40
30
20
10
16
10
22
II)
34
40
42
22
20
34
40
42
B lmx
SerumFe M-
40
30
20
10
L,
10
16
W&S
Fig. 2. A the COUTW of serum iron. M+, maternal blood with iron therapy; M--, maternal blood without iron therapy; -. mean value for each group (15 persons).
85
-,-34
4O
42
w66ks
86
h/l 125
i
A Fermn
M+
\ 1001
Ferrittn M-
25
-
22
28
24
Fig. 4. A time couree for semm ferritin. M+, maternal blood with iron therapy; M-, mean value for each group (16 persons). out iron therapy;-,
40
42
Weeks
maternal blood with-
ng/ml
25a
204
isa
.
ioa
.
SO
.
t :
A-
. 0
a#
!
M-
M+
.
F
c-
C+
Fig. 6. Ferritin concentration in maternal blood at term and cord blood, M+, matcmal blood with iron therapy; M-•, maternal blood without iron therapy; C+, cord blood with iron therapy; C-, cord blood without iron therapy; -, mean value.
.
mmol/ I 124
Hemoglobin
.
..
.
. 1O.c
_ T .
i. .. 8.0
.. * . . .. .
60
Fig. 6. Hemoglobin concentration Fie. 5).
t
‘
+. : . a. .
.
. M-
.. .
M+
c-
C+
in maternal blood at term and cord blood for M+, M-, c+ &
C--(ee
88 / umol/l
~ Serum Fe
40
. 30
. i !
‘: t
*
”
.. . :
20
T
. . . . .
. 10
: 1
.
I .
-f .
I
I
Ol Fig. 7. Serum iron concentration Fig. 5).
in maternal blood
at term and cord blood for M+. M-_, C+ and C- (See
wllol/ I Transferm
751
.
I
-L i
50
I
A
I
.
:
a.
.
-?i
..
25
M-
Fig. 8. The serum ferritin C- (see Fig. 6).
M+
concentration
cin maternal blood
c+ at term and cord blood
for M+, M--, C+ and
89 TABLE
I
Mean values (n = 15) + 1 S.D. Hb (mmol/l)
Tr (pmol/U
SeFe (pmolll)
Feaitin (ugll)
Maternal serum at term without Fe
1.0 + 1
64 ? 5
9?3
Maternal serum at term with Fe
0.2 * 0.1
52 ? 6
21 ? 5
Cord blood without Fe
9.7 f 1.8
33 f 5
24 + 5
168 + 44
Cord blood with Fe
9.7 + 0.0
32 + 4
22 f 6
174 f 46
TABLE
6+22f
6 12
II Hb (mmol/l)
? 1 S.D.
Ferritin
(ug/l)~+ 1 S.D.
Maternal 12 weeks after term without Fe during pregnancy
0.1 + 0.3
18?
9
Maternal 12 weeks after term with Fe during pregnancy
a.7 ? 0.3
23+
12
Maternal at term without Fe during pregnancy
1.0 f 1
5+
6
Maternal at term with Fe during pregnancy
8.2 + 0.7
22+
12
Some results, as an example five or six curves in one figure for each group of analysis, are shown in Figs. l-4. Also the mean curve for each group (15 persons) is shown in these figures. In Fig. 1 we observe a fall in Hb concentration until week 28, followed by a slight increase. In Fig. 2 a slight and gradual fall in serum’iron concentration during pregnancy is observed. Fig. 3 shows a rise in transferrin concentration which is in agreement with the fall in serum iron concentration as shown in Fig. 2. Fig. 4 shows a fall in ferritin concentration during pregnancy. The mean ferritin concentration at week 14 is 53 pg/l in group A and 47 pg/l in group B. Statistical analysis shows no significant differences in Hb, Tr and serum iron between group A and B at 14 and.28 weeks. At term, however, there are significant differences in the concentration of Hb, Tr, serum iron and serum ferritin between group A and B, as can be seen from Figs. 5-S. In Figs. 5-8 a comparison of values in blood at term and in cord blood between the groups A and B is shown. It is clear from the figures that there are no differences between cord blood values in group A and B. This means that the fetus does not discriminate relative to the iron status of the mother. The comparative data are also given in Tables I and II. Discussion The burden of pregnancy with its inevitable depletion of body iron commonly results in an adequate iron status becoming transformed to a state of
90
iron deficiency. If maternal stores do not exist, or cannot be mobilised, the total costs of pregancy, some 1000 mg, will have to be met from diet. If iron intake is inadequate the component which appears to suffer is the increase in maternal red cell mass. Starting with a normal mean value (Hb = 8.J mmol/l) for Hb in the thirty pregnant women we have seen a fall in Hb till X = 7.0 at term for women with an inadequate intake without any iron supplement, while women with an intake of 100 mg iron/day had at term a mean value of 8.2 mmof Hb/l. Parallel with this fall in Hb concentration is the decrease in serum iron concentration, in group A to 9 pmol/l. Starting with a mean serum ferritin concentration of about 50 pg/l (normal value for women) we observe a remarkable fall in ferritin concentration of about 30-40 pg/l. This suggests a demobilisation of the iron storage pool by the demands of normal pregnancy *. Literature data [2,3&J] suggest that 1 pg ferritin per litre of serum represents 8 mg of storage iron. In case of a normal value for the blood volume, a Ht of 0.5, a value for the serum volume of 2-2.5 1 is valid. Then a fall of 30-40 pg/l reflects a loss of iron of about 700 mg iron. A mean difference in serum ferritin at term between group A and B, namely 17 pg/l, reflects a difference in storage iron of about 275 mg, this is in agreement with a difference of 1.2 mmol Hb/l (8.2-7.0 mmol Hb * 285 mg Fe). Three months after pregnancy we determined once more the Hb concentration. For both groups the Hb has raised to normal value, 8.7 mmolfl. In contrast with the normal Hb value, surprisingly the serum ferritin concentration was still low. This should mean that for the restore of the Hb balance the iron need has come from liver ferritin. A low serum ferritin agrees well with lower iron storage in the liver. For therapeutic aims one should suggest from these data that a higher iron intake after pregnancy for some time is useful. How far the serum ferritin concentration reflects the iron content of the liver is difficult to prove exactly. In animal experiments we now follow during pregnancy the iron content of maternal and fetal liver, besides the maternal serum ferritin concentration. Results of this investigation, to be published elsewhere, confirm the usefulness of the determination of serum ferritin concentration. The iron status of the mother (group A and B) does not make any significant difference for the iron parameters of the fetus. In both groups there are equal values for Hb, serum iron, Tr and serum ferritin. So the fetus does not discriminate as to the iron status of the mother. References 1 Verhoef, NJ. and van Eijk, H.G. (1975) Clin. Sci. Mol. Med. 46.335-340 2 Jacobs, A. and Wonvood, M. (19743 Iron in Biochemistry, Chapters 11 and 13, Academic Press, New York 3 Walters, G.O., Miller, F.M. and Worwood. M. (1973) J. Clin. Pathol. 26, 770-772
* When this paper was submitted for publication a paper titled “Ferritin as an assessment of iron stores in normal pregnancy” by A.M. Kelly et al. appeared in the British Journal of Obstetrics and Gynaecology 84,1977.434438 (June).
91 4 Wiltink. W.F.. Kruithof. J.. Mol, C.. Bos. G. and van Eijk, H.G. (1973) 5 Halliday. J.W.. Gera. K.L. and Powell, L.W. (1976)
Clin. Chim. Acta 49.99-104
Clin. Chim. Acta 53.207-214
6 van Eijk, H.G. (1976) Austr. Family Physician 6.52-66 7 Sturgeon, P. (1969) Br. J. Haematol. 5,3144.46-66 3 Rios. E.. Lips&its. D.A., Cook, J.D. and Smith, NJ. (1976)
Pediatrics 55.694-699
Clinica Chimico Acta, 83 (1978) @ Elsevier/North-Holland
81-91
Biomedical
Press
CCA 9048
SERUM FERRITIN
H.G. VAN
AND IRON STORES DURING PREGNANCY
EIJK ‘+*, M.J. KROOS
*, G.A. HOOGENDOORN
b and H.C.S. WALLENBURG
b
a Department of Chemical Pathology and b Department of Gynaecology and Obstetrics, Medical Faculty, Erasmus University Rotterdam, Rotteniam (The Netherlands) (Received
July 15th,
1977)
summary The iron status of two groups of pregnant women was investigated. One group did not receive iron (group B), the other received 100 mg iron/day (group A). 1. In all individuals concentrations of hemoglobin, serum iron, transferrin and serum ferritin were determined at regular intervals from the third month until delivery and at 3 months after delivery. The same determinations were performed in cord blood. 2. Changes in iron status appeared to be less in individuals with iron supplement than in those without iron supplement. A fall in Hb, serum iron and serum ferritin is observed in all individuals. 3. Three months after delivery the Hb concentration has generally returned to the normal female value, but the serum ferritin concentration is still very low. 4. The fetus does not discriminate as to the iron status of the mother. In both groups (A and B) cord blood values appeared to be not significantly different.
Introduction Ferritin is the main storage compound for iron in the human body and is present mainly intracellularly in the reticuloendothelial system of the liver, the spleen and the bone marrow. Intracellular ferritin can be identified in histologic sections of bone marrow smears by staining with Prussian blue. A small amount is normally found in circulating plasma in concentration varying between 10 and 200 pg/l. In men the serum concentration is twice that of women, suggesting that the serum level reflects total body stores (liver, spleen, bone marrow, *
To whom
correspondence
should be addxeued.
82
Hb). In case of iron deficiency or iron overload the serum ferritin concentration is related to the concentration in the tissues. It should be noted, however, that in patients with tissue damage (acute and chronic liver disease) or malignancy (leukemia) the relationship with the iron status may be disturbed by variations in abnormal production or release of ferritin. During pregnancy an appreciable amount of iron is transferred from the mother to the fetus. At term, the human fetus contains 200-400 mg iron. The mechanism by which iron is transferred from the mother to the fetus across the placenta is poorly understood, notwithstanding a number of investigations in various animals [ 11. It is likely that the mobile iron pool in the liver is involved in the mobilization of iron. The object of the present study was to evaluate the significance of the determination of serum ferritin as a measure of the iron reserves during pregnancy and, therefore, as an indication for the application of iron therapy. A practical advantage of the radioimmunoassay of serum ferritin is that it provides a virtually noninvasive method for measuring a segment of the body ferritin pool [ 2,3]. Materials and methods Thirty pregnant women, attending the University Hospital/Obstetrical Clinic, were randomly selected for this study. The course of pregnancy and delivery was uncomplicated in all cases. Fifteen women (group A) received 100 mg iron daily (as FeSOj,) from the third month of pregnancy until delivery. Fifteen women did not receive any iron supplement (group B). Venous blood samples were obtained between 14.00 and 16.00 hour every three or four weeks from the third month until delivery and at three months after delivery. Five ml samples were collected in polystyrene tubes containing Sequestrene@ as an anticoagulant. Mixed cord blood was obtained immediately after delivery. The hemoglobin concentration was determined with a Coulter Counter Model 5, The inst~ment was calibrated with hematology reference control 4C (Coulter Diagnostics Inc., Hialek, Fla.). Serum iron and transferrin concentrations were determined as described previously [ 41. For the determination of serum ferritin a specific highly-sensitive solid phase radioimmunoassay was used, adopted from Ha&day [ 51, as reported earlier in detail [ 61. The reproducibility of all methods used was more than 95%. Identical variables from the groups A and B were statistically compared using the Wilcoxon test (two-tailed). Results We investigated the Hb, Tr, serum iron and serum ferritin concentrations in maternal blood at term, in maternal blood twelve weeks after term (supposing that the Hb values in this blood would be equal to those in the blood at time zero, before pregnancy), in cord blood at term and a time-course of I-Ib, Tr, serum iron and serum ferritin in about 15 pregnent women with (group A) and 15 women without (group 3) any iron supplement during their six-month follow-up period until delivery.
83
A mmot/l
Hemoglobin
M*
104
9.0
7’0
I 40
42
40
42
Weeks
B
mmol/l
Hemoglobin
10
16
M-
22
2a
34
Fig. 1. A time course of hemoglobin. M+. maternal blood with iron therapy; M-, iron therapy; -, mean value for each group (15 persons).
#oks
maternal blood without
84
(A Serum Fe
M+
40
30
20
10
16
10
22
II)
34
40
42
22
20
34
40
42
B lmx
SerumFe M-
40
30
20
10
L,
10
16
W&S
Fig. 2. A the COUTW of serum iron. M+, maternal blood with iron therapy; M--, maternal blood without iron therapy; -. mean value for each group (15 persons).
85
-,-34
4O
42
w66ks
86
h/l 125
i
A Fermn
M+
\ 1001
Ferrittn M-
25
-
22
28
24
Fig. 4. A time couree for semm ferritin. M+, maternal blood with iron therapy; M-, mean value for each group (16 persons). out iron therapy;-,
40
42
Weeks
maternal blood with-
ng/ml
25a
204
isa
.
ioa
.
SO
.
t :
A-
. 0
a#
!
M-
M+
.
F
c-
C+
Fig. 6. Ferritin concentration in maternal blood at term and cord blood, M+, matcmal blood with iron therapy; M-•, maternal blood without iron therapy; C+, cord blood with iron therapy; C-, cord blood without iron therapy; -, mean value.
.
mmol/ I 124
Hemoglobin
.
..
.
. 1O.c
_ T .
i. .. 8.0
.. * . . .. .
60
Fig. 6. Hemoglobin concentration Fie. 5).
t
‘
+. : . a. .
.
. M-
.. .
M+
c-
C+
in maternal blood at term and cord blood for M+, M-, c+ &
C--(ee
88 / umol/l
~ Serum Fe
40
. 30
. i !
‘: t
*
”
.. . :
20
T
. . . . .
. 10
: 1
.
I .
-f .
I
I
Ol Fig. 7. Serum iron concentration Fig. 5).
in maternal blood
at term and cord blood for M+. M-_, C+ and C- (See
wllol/ I Transferm
751
.
I
-L i
50
I
A
I
.
:
a.
.
-?i
..
25
M-
Fig. 8. The serum ferritin C- (see Fig. 6).
M+
concentration
cin maternal blood
c+ at term and cord blood
for M+, M--, C+ and
89 TABLE
I
Mean values (n = 15) + 1 S.D. Hb (mmol/l)
Tr (pmol/U
SeFe (pmolll)
Feaitin (ugll)
Maternal serum at term without Fe
1.0 + 1
64 ? 5
9?3
Maternal serum at term with Fe
0.2 * 0.1
52 ? 6
21 ? 5
Cord blood without Fe
9.7 f 1.8
33 f 5
24 + 5
168 + 44
Cord blood with Fe
9.7 + 0.0
32 + 4
22 f 6
174 f 46
TABLE
6+22f
6 12
II Hb (mmol/l)
? 1 S.D.
Ferritin
(ug/l)~+ 1 S.D.
Maternal 12 weeks after term without Fe during pregnancy
0.1 + 0.3
18?
9
Maternal 12 weeks after term with Fe during pregnancy
a.7 ? 0.3
23+
12
Maternal at term without Fe during pregnancy
1.0 f 1
5+
6
Maternal at term with Fe during pregnancy
8.2 + 0.7
22+
12
Some results, as an example five or six curves in one figure for each group of analysis, are shown in Figs. l-4. Also the mean curve for each group (15 persons) is shown in these figures. In Fig. 1 we observe a fall in Hb concentration until week 28, followed by a slight increase. In Fig. 2 a slight and gradual fall in serum’iron concentration during pregnancy is observed. Fig. 3 shows a rise in transferrin concentration which is in agreement with the fall in serum iron concentration as shown in Fig. 2. Fig. 4 shows a fall in ferritin concentration during pregnancy. The mean ferritin concentration at week 14 is 53 pg/l in group A and 47 pg/l in group B. Statistical analysis shows no significant differences in Hb, Tr and serum iron between group A and B at 14 and.28 weeks. At term, however, there are significant differences in the concentration of Hb, Tr, serum iron and serum ferritin between group A and B, as can be seen from Figs. 5-S. In Figs. 5-8 a comparison of values in blood at term and in cord blood between the groups A and B is shown. It is clear from the figures that there are no differences between cord blood values in group A and B. This means that the fetus does not discriminate relative to the iron status of the mother. The comparative data are also given in Tables I and II. Discussion The burden of pregnancy with its inevitable depletion of body iron commonly results in an adequate iron status becoming transformed to a state of
90
iron deficiency. If maternal stores do not exist, or cannot be mobilised, the total costs of pregancy, some 1000 mg, will have to be met from diet. If iron intake is inadequate the component which appears to suffer is the increase in maternal red cell mass. Starting with a normal mean value (Hb = 8.J mmol/l) for Hb in the thirty pregnant women we have seen a fall in Hb till X = 7.0 at term for women with an inadequate intake without any iron supplement, while women with an intake of 100 mg iron/day had at term a mean value of 8.2 mmof Hb/l. Parallel with this fall in Hb concentration is the decrease in serum iron concentration, in group A to 9 pmol/l. Starting with a mean serum ferritin concentration of about 50 pg/l (normal value for women) we observe a remarkable fall in ferritin concentration of about 30-40 pg/l. This suggests a demobilisation of the iron storage pool by the demands of normal pregnancy *. Literature data [2,3&J] suggest that 1 pg ferritin per litre of serum represents 8 mg of storage iron. In case of a normal value for the blood volume, a Ht of 0.5, a value for the serum volume of 2-2.5 1 is valid. Then a fall of 30-40 pg/l reflects a loss of iron of about 700 mg iron. A mean difference in serum ferritin at term between group A and B, namely 17 pg/l, reflects a difference in storage iron of about 275 mg, this is in agreement with a difference of 1.2 mmol Hb/l (8.2-7.0 mmol Hb * 285 mg Fe). Three months after pregnancy we determined once more the Hb concentration. For both groups the Hb has raised to normal value, 8.7 mmolfl. In contrast with the normal Hb value, surprisingly the serum ferritin concentration was still low. This should mean that for the restore of the Hb balance the iron need has come from liver ferritin. A low serum ferritin agrees well with lower iron storage in the liver. For therapeutic aims one should suggest from these data that a higher iron intake after pregnancy for some time is useful. How far the serum ferritin concentration reflects the iron content of the liver is difficult to prove exactly. In animal experiments we now follow during pregnancy the iron content of maternal and fetal liver, besides the maternal serum ferritin concentration. Results of this investigation, to be published elsewhere, confirm the usefulness of the determination of serum ferritin concentration. The iron status of the mother (group A and B) does not make any significant difference for the iron parameters of the fetus. In both groups there are equal values for Hb, serum iron, Tr and serum ferritin. So the fetus does not discriminate as to the iron status of the mother. References 1 Verhoef, NJ. and van Eijk, H.G. (1975) Clin. Sci. Mol. Med. 46.335-340 2 Jacobs, A. and Wonvood, M. (19743 Iron in Biochemistry, Chapters 11 and 13, Academic Press, New York 3 Walters, G.O., Miller, F.M. and Worwood. M. (1973) J. Clin. Pathol. 26, 770-772
* When this paper was submitted for publication a paper titled “Ferritin as an assessment of iron stores in normal pregnancy” by A.M. Kelly et al. appeared in the British Journal of Obstetrics and Gynaecology 84,1977.434438 (June).
91 4 Wiltink. W.F.. Kruithof. J.. Mol, C.. Bos. G. and van Eijk, H.G. (1973) 5 Halliday. J.W.. Gera. K.L. and Powell, L.W. (1976)
Clin. Chim. Acta 49.99-104
Clin. Chim. Acta 53.207-214
6 van Eijk, H.G. (1976) Austr. Family Physician 6.52-66 7 Sturgeon, P. (1969) Br. J. Haematol. 5,3144.46-66 3 Rios. E.. Lips&its. D.A., Cook, J.D. and Smith, NJ. (1976)
Pediatrics 55.694-699
