Professional Documents
Culture Documents
com
POSTGRAD. MED. J. (1962), 38, 202
ANAEMIA IN PREGNANCY
JEAN M. SCOTT, M.D.
Assistant Pathologist, Research Department, Royal Maternity Hospital, Rottenrow, Glasgow.
THOUGH one normally regards the estimation of process. It appears towards the end of the first
hemoglobin as one of the simplest procedures, it trimester, increases more rapidly thereafter and
does seem rather anomalous that having obtained usually reaches a maximum around the thirty-sixth
the result, it should then be difficult to decide week. Towards term some slight reduction in
whether certain pregnant women are actually plasma volume has been reported (Lund, 1951;
anemic or not. This is largely due to other Cope, 1958). Thus the so-called threshold of
changes which take place in the blood during anaemia in pregnancy is a variable one and many
pregnancy. As early as I88I Willcocks first noted authors have noted the falling curve of hemoglobin
the occurrence of hydraemia. Since then more levels throughout gestation and the slight rise just
detailed investigations have been carried out before term (Evans, 1943; Elliott, 1944; Scott and
(Dieckmann and Wegner, I934; Thomson, Hir- Govan, 1949; Rath, Caton, Reid, Finch and
sheimer, Gibson and Evans, I938; Roscoe and Conroy, 1950; Giles and Burton, I960).
Donaldson, I946; Tysoe and Lowenstein I950; Most of these surveys have covered large
Lund and Sisson, 1958), which have shown that numbers of pregnant women and a fairly regular
while all elements of the blood increase in decline in the average hemoglobin values is noted.
pregnancy, the incrementForin plasma Dieckmann
volume is More detailed analysis of individual patients,
proportionately greater. example
and Wegner (I934) estimated that the plasma
however, reveals a more variable picture. For
volume increased by 25%, and the red cell and
example Lund (I951) showed that the percentage
increase in plasma volume varied from 14% in
haemoglobin mass by only 20 and 15% respectively, one patient to as much as 121% in another and in
findings which explain the apparent drop in the many of their cases Roscoe and Donaldson (1946)
level noted by many other observers
haemoglobin Fullerton and Campbell, 1935; Boy-
found the blood volume increase was not outside
(Davidson, the range of non-pregnant controls. Even when
cott, 1936; Widdowson, I939; Davidson, Donald- the blood volume was related to weight or surface
son, Lindsay and McSorley, 1943; Reid and area and expressed in absolute terms, wide
Mackintosh, 1943; Elliott, I944; Fullerton, Mair variations were still present.
and Unsworth, I944) and called the' physiological Obviously, if one speaks of a lowered threshold
anemia' of pregnancy. of anaemia in pregnancy one must use the term in
In a recent review of the literature Kerr and its broadest sense, realising that it cannot be
Davidson (1958) tabulated the salient features of applied with any great accuracy to individual
this so-called physiological state, comparing and patients. This was generally acknowledged and
contrasting them with those of true iron deficiency up till recently the idea of a physiological anaemia
anaemia. They found that a normal colour index, of pregnancy was accepted. It was then noted that
low-normal MCHC, normal or low-normal serum iron therapy could correct the fall in haemoglobin
iron and low-serum-iron/iron-binding-capacity and even prevent its occurrence (Scott and Govan,
ratio, slightly raised reticulocyte count and I949; Klopper and Ventura, 1951; Benstead and
slightly increased free erythrocyte protoporphyrin
had all been reported. Most authorities agreed
Theobald, I952; Davis and Jennison, I954;
Lillie, Gatenby and Moore, 1954; Fisher and
that these findings were unlikely in a true anaemia Biggs, 1955; Edgarevidence
and Rice, 1956; Kerr and
and they attributed the lowered hemoglobin level Davidson, 1958), strongly against any
to haemodilution. Adair, Dieckmann and Grant so-called physiological process. Large numbers
(1936) considered that only haemoglobin values of otherwise normal pregnant women were
below 10 g./Ioo ml. could be considered abnormal, followed throughout gestation, no anemias being
and Whitby and Britton (i953) and many others included. Some were given iron therapy and the
with this. Holly (1953) on the other hand
agreed the others left untreated or given a placebo. Not only
placed level higher at ii.5 g./Ioo ml. did iron therapy eliminate the typical falling curve
The problem is not simply defined, however, of hemoglobin levels, but in treated cases the
largely because the hydraemia is a progressive haemoglobin reading at term was higher than that
Downloaded from http://pmj.bmj.com/ on September 26, 2017 - Published by group.bmj.com
found initially. Even when treatment was delayed Allaire and Campagna (I96I) found normal iron
till the twenty-fourth week Benstead and Theobald storage in four patients with hemoglobin values
(I952) and Kerr and Davidson (1958) showed that between 7.6 and io g./ioo ml. Unfortunately
the developing 'physiological anaemia' could be blood volume estimations were not made on these
corrected and they recommended that all pregnant patients, but such a combined study might prove
women should be given iron from the twenty- enlightening.
Iron stores are usually reduced in pregnancy
fourth week onwards. More recently Giles and
Burton (1960) stated that 'the hemoglobin of a (Holly and Grund, I959; Allaire and Campagna,
normal woman in late pregnancy should be at 1961). Once iron therapy is instituted however,
least 80% (II.8 g./Ioo ml.)'. stainable iron begins to appear in the marrow and
Possiblyensome reservation ought to be made here. the majority of patients show a rise in hemoglobin.
Surveys masse, as above, are not concerned These authors regard this as further evidence
with the treatment of patients as individuals. that iron deficiency exists. According to them one
They deal merely with hemoglobin levels, which should aim at maintaining normal iron stores
are purely relative, for as Benstead and Theobald throughout pregnancy, irrespective of the increase
(I952) pointed out, no one really knows what in total haemoglobin mass. In hypervolaemic
happens to the blood volume when the hemoglobinin patients this would entail the uptake of a consider-
begins to rise. Moreover the rise is not equal able amount of iron, if it were possible, and might
all patients. Scott and Govan (I949) divided their lead to iron overload in the puerperium, when the
cases into three groups, 40% who improved, blood volume returns to normal and the mother
52% who showed no change and 8% who receives a ' pay-back' of at least 500 mg. of iron
deteriorated. Hamilton and Wright (1942) from the reduction in her own hemoglobin mass
showed similar findings, but more patients and others, I950). According to Hahn
(Rath this
deteriorated. Other reviews mention anything (1937) amount of iron would represent a 50%
from a io to 20% failure-rate with patients still increase above normal in available reserves.
anaemic at term (Fisher and Biggs, I955; Edgar Obviously this problem requires further investi-
to the British Medical Journal
and Rice, I956); and Kerr and Davidson (I958) gation. there
According
attributed these to either (a) a failure to take iron, is now strong evidence that excessive
(I958) of haemosiderin cannot produce hemo-
(b) failure toor absorba it, (c)increase
a mild folic acid deposits
deficiency, (d) large in plasma chromatosis (Cappell, Hutchison and Jowett,
volume. A number of patients may be included 1957). The only effects of excessive accumulation
of iron might be the relative deficit of tissue
in the first two categories, but the remainder are
not explained wholly by the third since Giles and antioxidants, especially vitamin E (Sisson and
Burton (1960) had failures even in their folic acid Lund, I957).
and iron-treated series. These authors noted that Until more is known about blood viscosity and
in at least 20% of the patients who remained iron storage in pregnancy, one wonders whether
'anamic' despite treatment, there was a sharp the maintainance of a haemoglobin level of over
rise in hemoglobin after delivery and they felt all pregnant women (Giles and Burton,
80% inis the
tempted to attribute the persistently low ante- I960) ideal and whether we should strive to
achieve this. Certainly there is no doubt that
partum values to hypervolaemia.
This certainly seems reasonable. In a survey many pregnant British women have a latent iron
covering 132 women throughout gestation, Lund deficiency and compared with their Australian
and Sisson (I958) found as many as 20% were sisters are in a poorer state of iron nutrition
actually hypervolemic. To raise or maintain (Morgan, 1961). On the other hand if routine
their haemoglobin levels would mean, of necessity, iron therapy is not instituted, it is difficult in a
a tremendous increase in total hemoglobin mass busy antenatal clinic to recognize early stores iron-
and could lead to further complications, for deficiency states. Assessment of body iron
Hamilton (1950) has shown that the relative and blood volume estimations, though helpful in
decrease in cellular elements, secondary to the odd case, cannot be regarded as routine
hydramia, asproduces a fall in blood viscosity. procedures. Serum iron levels and even MCHC
Moreover, the viscosity fell the cardiac output values are not good indicators of iron store
rose. By this means she felt some mechanical
relief was given to the heart, and, it may be, that
depletion (Allaire and Campagna, I96I) though a
rise in the former and a fall in the total iron-
the persistently low hemoglobin values in binding capacity with iron therapy can be taken
hypervolemic patients represent some such pro-
tective mechanism and are not the result of
as biochemical evidence of an iron deficiency
state (Morgan, I96I). Neither of these methods
failure to respond. would however be recommended as a routine test
In this respect it is also interesting to note that and both entail repeat estimations. One must
Downloaded from http://pmj.bmj.com/ on September 26, 2017 - Published by group.bmj.com
gave their conclusion 'that its use in the clinical Megaloblastic Anemias
dosage recommended carries a negligible risk and
is probably less hazardous in other respects than
The earliest report on this severe ' anhemia'
of pregnancy was made in i842 by Channing,
intravenous iron preparations and blood transfu- an American obstetrician. Though occasional case
sions' (Duthie, Girdwood, Hubble, MacGregor, reports appeared in the literature thereafter, no
Wayne, Wilson and Wilson, I960). Since the real survey was made until Osler published his
publication of this letter most centres are now paper in I9I9. He noted that, unlike Addisonian
using iron-dextran complex, but as a later British
Medical Journal Editorial (Ig6ob) and Haddow pernicious anemia, recovery if it took place was
(1960)risk
pointed out, the possibility of any carciono- permanent and the woman might escape the
anaemia in subsequent pregnancies. Recovery in
genic will not be apparent until another 5 years
have passed.
those days before transfusion, liver or folic acid
Within the last few months a new intra- therapy was by no means certain and the maternal
muscular haematinic complex of iron sorbitol and mortality and stillbirth rates were high (Balfour,
citric acid with dextran has become available in 1927; Wintrobe, 1942).
this country (Jectofer-Lindvall and Andersson, While still regarded as a rarity, reports on ever-
I96I). The molecule is smaller than that of iron increasing numbers of cases began to appear in
the literature. Stevenson (1938) described 30
dextran (Imferon) and is absorbed directly into
the blood stream from the injection site, reaching cases in a 6-year survey, Davidson (I95 ) reported
maximum values there within two to three hours on 42 cases in i years, Thomson and Ungley
(Andersson, 1961): 30% of the injected dose is found 46 cases in 17 years, and Lund
(1951) estimated
lost in the urine and to allow for this a greater (I95I) that I% of all anemias were of
amount of iron would have to be injected. Un- this type. More recently the incidence has been
fortunately the molecule is less stable than that expressed as a percentage of total deliveries and
of iron dextran and the makers advise a maximum for comparison figures from various centres are
single dose of o00 mg. The number of injections listed in Table 2.
has therefore to be increased, but so far no It can be seen that even in the last 7 years the
reactions of any import have been reported and incidence has been rising. It may be that this is
the main advantage is that to date no carcinogenic due to a better understanding of the anemia and
effect has been noted in animals (Andersson, earlier recognition. In Glasgow there is a slight
Lundin and Lundin, I96i). Trials are continuing, rise between I954 and I959, but this is not
however, in several centres. statistically significant (standard error 0.155). The
With intramuscular iron dextran the average figures include only cases of frank megaloblastic
case will show a rise in haemoglobin of just under anaemia, whereas transitional forms of the anemia
I g./ioo ml. in one week. Where labour is im- are included in some of the other reports. This
minent and the anaemia severe, one must decide may partly account for the higher incidence
what improvement can be expected and whether elsewhere.
transfusion is necessary. In this centre we aim to A seasonal variation has also been reported, a
have the haemoglobin over 9 g./Ioo ml. at least higher incidence being found in late winter and
since our findings show that the risk to the mother springand
(Holly, 1951; Forshaw, 1957; Thompson,
above this level is probably no greater than the 1957) it has been related to the lower folic acid
risk of transfusion itself (Brit. Med. J., 1955; levels in the diet at this time of the year (Thomp-
Bare and Sullivan, I960). son, 1957). No seasonal pattern was found by
Downloaded from http://pmj.bmj.com/ on September 26, 2017 - Published by group.bmj.com
..
4.3 (2.8)
0.9 ()
3.0 (2.9)
6.0 (1.5)
-
-
.6 (o.) f 4.7 (5.-)
5.1(2.1)
Premature labour .. .. - 9.3 (I2.5) - I. 42.0(13.2)
Prolonged labour .. .. 1.0 (17) - 7.5 (22.0)
(more than (more than
48 hrs.) 26 hrs.)
Forceps delivery .. .. 10.0 (13.0) - 21.0 (II.o)
P.P.H. .. .. .. 3.4 (5.6) 2.0 (22.0) -- II. (5.0)
Miscarriage ....... . I I.0
*
Figures in brackets represent the incidence of complications in the ordinary hospital population.
t Figures in brackets represent the incidence of complications in ' low normal ' (Hb. Io-II g./ioo ml.) controls.
t Figures in brackets represent the incidence of complications in all patients with Hb. values over 9 g./Ioo ml.
multiple pregnancy, with its higher incidence of before delivery, if not fully investigated haemato-
megaloblastic anaemia. logically.
Further reference to Table 3 will show that
three of the surveys (Gatenby and Lillie, I960;
Complications of Anemia in General McKenzie and Abbot, I960; Ainley, I96I) deal
Before the introduction of safe blood transfusion solely with megaloblastic anaemia of pregnancy.
and specific anti-anamic therapy the prognosis for While the complications are generally similar to
both mother and child was poor. Recently no those mentioned above this type of anaemia
maternal deaths have been recorded in this appears to be associated with a higher incidence
country, but even here one cannot afford to be of toxaemia, accidental haemorrhage and possibly
complacent about the incidence of puerperal hydramnios. These relationships have only
the significantly raised miscarriage,
morbidity or and recently become apparent, as more cases of this
prematuritywhere perinatal mortality rates. In anaemia come to light, but the underlying condi-
Mombasa, anaemia is prevalent, the full tions responsible for their occurrence remain as
force of its effect is apparent (MacGregor, I96I). yet obscure. Whether or not there is a link with
These conditions exist today, but the cases folic acid or other deficiencies, there can be no
analysed in this department were actually drawn doubt of the value of early diagnosis and treatment
from the initial years of our survey (1946 and of anaemia in pregnancy and of the considerable
1947), when follow-up methods at the ante-natal part this has played in recent years in the
blood clinic were not so well organized. Today all reduction of maternal morbidity and fcetal
these patients would at least have been transfused perinatal mortality.
REFERENCES
ABRAMS, J. (I959): Sickle Cell Trait and Pregnancy, Obstet. Gynec., I4, 123.
, and SCHWARTZ, I. R. (I959): Sickle Cell Diseases in Pregnancy, Amer. J. Obstet. Gynec., 77, I324.
ABRAMSON, L. (1938): Etats rappelant l'anemie pernicieuse durant la grossesse et les suites de couches, Acta med.
scand., 96, 319.
ADAIR, F. L., DIECKMANN, W. J., and GRANT, K. (1936): Anemia in Pregnancy, Amer. J. Obstet. Gynec., 32, 50.
ADAMS, J. F. (1958): Pregnancy and Addisonian Pernicious Anemia, Scot. med. J., 3, 21.
AGUERO, 0., and LAYRISSE, M. (1958): Megaloblastic Anemia of Pregnancy in Venezuela, Amer. J. Obstet. Gynec.,
76, 903.
AINLEY, N. J. (1961): Megaloblastic Anaem a of Pregnancy and the Puerperium, J. Obstet. Gyncec. Brit. Emp., 68, 254.
Downloaded from http://pmj.bmj.com/ on September 26, 2017 - Published by group.bmj.com
Anmia in Pregnancy
Jean M. Scott
These include:
Email alerting Receive free email alerts when new articles cite
service this article. Sign up in the box at the top right
corner of the online article.
Notes