Downloaded from by guest on June 11, 2014 ajcn.nutrition.org

Folacin, cobalamin,
and hematological
status
during pregnancy
in rural Kenya: the influence
parity, gestation,
and Plasmodium
falciparum
malaria1
BJ Brabin,
MSc,
PhD,
FRCP(C),
H van den Berg,
PhD,
of
and F Nzjmeyer
To investigate
folacin concentrations
in malaria during pregnancy,
women attending
a rural antenatal
clinic in Kenya were studied. Low serum folacin values had poor specificity
for low
red blood cell (RBC) folacin concentrations.
Multigravidae
had lower mean serum folacin (p <0.03)
and RBC folacin (p <0.001)
values than primigravidae.
Primigravidae
had higher mean RBC folacin
values than nulliparae
(p <0.05).
Although anemia was frequent,
no evidence of neutrophil
hypessegmentation was seen in blood smears of individuals
with low RBC folacin or indeterminate
cobalamin
values. The unexpectedly
high RBC folacin concentrations
are probably
related to Pfalciparum
infection:
during followup
a significant
decrease
in both RBC and serum folacin activity occurred
after chioroquine
was administered.
This decrease
may be unrelated
to a gestational
effect (RBC
folacin p <0.01; serum folacin p <0.025).
The pathogenesis
of high RBC folacin activity is discussed
in relation to reticulocytosis
as well as to a biochemical
mechanism within the RBC.
Am I Clin
Nuir 1986;43:803-8l5.
ABSTRACF
WORDS
Folic
acid, cobalamin,
Introduction
Previous studies have reported the incidence
of megaloblastic
anemia in pregnancy
in several countries
in Africa (1-5). All of these reports are from urban areas and concern
hospital admissions
or attendances.
Few report
serial changes during pregnancy
of serum folacin or cobalamin
values (6, 7). The literature
provides
little information
on red blood cell
(RBC) folacin and none on the influence
of
malaria
on RBC folacin in pregnancy.
The effect of parity on the pattern of hematological
changes during pregnancy
has been poorly defined.
Recent studies of the anemia
of Pfakiparum infection
in children
report that high RBC
folacin values occur in malaria (8, 9). Interestingly,
it is the red cell and not the serum
folacin values that are raised in these children,
and for those with chronic anemia the reported
mean value is above the upper limit of the
normal
assay range (8).
The aim of the present study was to define
the hematological
status of a rural pregnant
population
in relation to parity, gestation,
and
The American
Journal of ClinicalNutrition43: MAY
© 1986 American
Society for Clinical Nutrition
1986,
pregnancy,
malaria,
anemia,
parity
Pfakiparum
infections.
It had been hypothesized in a previous
study that folacin deficiency,
if considered
as a common
complication of malaria in pregnancy,
would lead to
maternal
immunosuppression
(10).
Materials
Location
and methods
and diet
The study was undertaken
in Western Province, Kenya,
amongst
women of the Abasamia
tribe, at an antenatal
clinic in a rural hospital at Nangina. The hospital is in the
Busia District, an area holoendemic
for malaria. The Abasamia are a subtribe of the Abaluyha
who live close to the
I
From
the Department
of Tropical
Hygiene
(BJB),
Royal Tropical
Institute, Amsterdam,
The Netherlands;
and Department
of Clinical
Biochemistry
(HVDB,
FN),
Central
Institutefor Nutrition
Research
(CIVO/TNO),
Zeist, The Netherlands.
2Suppo
by a project grant from the Medical
Research Centre, a department
of the Kenya Medical
Research Institute, Nairobi.
3Address reprint requests to: BJ Brabin, PhD, Institute
of Medical Research, P0 Box 378, Madang, Papua, New
Guinea.
Received
August 20, 1984.
Accepted
for publication
June 25, 1985.
pp 803-8
15. Printed
in USA
803
Downloaded from ajcn.nutrition.org by guest on June 11, 2014
KEY
BRABIN
804
Lake Victoria shoreline.
The lake area has seen a flow of
migration
to Nairobi, for the agricUltUral potential of much
of its land is marginal
and Lake Victoria’s
waters have
been overfished.
The women cultivate maize asthe primary
subsistence
food, and cassava is the secondary
staple for
the preharvest
period. Meat, although
available, is too cxpensive for most women to buy on a regular basis. Fish is
less expensive
and is brought daily to the local markets.
Participants
and follow-up
Clinic
procedure
Each time the women attended
the clinic, they were
given antenatal
care by the nursing staff of the hospital.
Maternal parity and date of last menstrual
period were
known in nearly all cases. Gestational
age, calculated
from
the last menstrual period, was checked against assessment
of fundal height. At the monthly antenatal
visit, 310 mg
chloroquine
(base) was administered
for malaria prophylaxis. A weekly dose would be the usual and better procedure but, accordingto
the hospital Public Health Team,
such a schedule would be difficult in this rural area. Problems include attendances,
regularity
oftaking
prophylaxis
as instructed,
cost, and distribution
and supply of the
medicine.
Whereas
this chloroquine
schedule was established hospital
policy and its effect was unknown,
no
change was requested
specifically
for our study. Women
were observed
to swallow their tablets. A 2-wk course of
ferrous sulfate (200 mg daily) was offered to all women
on their first visit. A therapeutic
course ofiron was offered
to women
only after referral to the hospital doctor. After
clinic attendance,
the women visited the laboratory.
The
purpose and nature ofthe study was explained
in the local
language and permission
was requested
for their participation and for taking a blood sample
by venipuncture.
Participants
were given a bottle containing
a 10% formolsaline solution and were requested
to return it containing
a fresh stool sample at their next clinic visit
Bleeding,
laboratory
procedure,
and storage
of samples
Blood (5 ml) was collected by aseptic venipuncture
at
each clinic visit. Half of this sample was centrifuged
and
the plasma stored at -12#{176}Cwithin 2 h. The remaining
2.5 ml of blood was first used for preparing
three thick
and one thin smear for microscopy,
and then placed in
an EDTA anticoagulant
bottle.
Blood
smears were stained
by Giemsa’s stain and examined
for malaria.
Additional
examination
of peripheral
blood films from pregnant
women subsequently
shown to have low RBC folacin or
indeterminate
serum cobalamin
concentrations
was carried
out by an experienced
technician
at the Department
of
Hematology,
Academic
Medical
Centre,
Amsterdam.
Neutrophil
hypersegmentation
was defined using the segmentation
index, which is a count ofthe number of lobes
in 100 neutrophils. Normal subjects have 200-300
lobes
with this count (segmentation
index 200-300).
Hemoglobin (Hb) concentration
was measured
by the cyanomethemoglobin
method
and the hematocrit
(Hct) by the
Hawksley
microhematocrit
(Belman
Hawksley,
Brackmills, Northampton,
England).
Stool samples were cxamined using a formol-saline ether sedimentation
method.
Within 2 h ofcollection,
a whole blood hemolysate
was
prepared
from the EDTA solution to a 1:20 dilution using
a 0.2% ascorbic acid solution. The hemolysate
was allowed
to stand, protected
from light, for
1 h at room temperature to permit the hydrolysis
of polyglutamates
by endogenous conjugase. The hemolysates were stored within
3 h of collection
at -12#{176}Cfor 1-2 wk and then subsequently at -20#{176}C.Plasma and hemolysate
samples were
forwarded
frozen to the Central Institute for Nutrition
Research
(CIVO/TNO),
Zeist.
Radioassay
Serum and RBC folacin were measured
by radioassay
(Becton-Dickinson
Simul TRAC, Becton-Dickinson
Immunodiagnostics,
Orangeburg,
NY, vitamin
B12 [57cjfolate [1251]radioassay
kit) using a folacin-binding
milk
protein (11, 12). Endogenous
folacin-binding
proteins that
have been observed in the sera of some pregnant
women
(13) were denatured
by an initial boiling step(14).
Serum
vitamin B12 was measured
as true rather than total cobalamin
by using a nonintrmnsic
factor
blocking
agent
(15, 16).
Statistical
analysis
A t test
differences
correction
Friedman
was used to assess the statistical significance
of
between mean assay values,
x2 test with Yates
for differences
between
proportions,
and the
test for nonparametric
analysis of variance.
Results
Study
sample
Pregnant
women,
1 19 primigravidae
and
172 multigravidae,
were studied at the time of
their first antenatal
visit; 19 pnmigravidae
and
55 multigravidae
made only a first visit. Blood
samples
for analysis
were available
from 78
primigravidae
and 108 multigravidae
at day
28, from 5 1 primigravidae
and 37 multigravidae at day 56. In nearly all cases, women
returned within a day oftheir
scheduled
followuptime.
Stool samples were obtained
from 89
primigravidae
and 67 multigravidae.
The large
majority
of women
in this study were either
asymptomatic
or complained
of only minor
symptoms.
Only one woman
with severe ane-
Downloaded from ajcn.nutrition.org by guest on June 11, 2014
The study period extended
from late January
to the
end ofApril,
1981, and covered the last 7 wk ofthe dry,
and first 6 wk of the wet season. The study protocol was
approved by a research committee
at the Royal Tropical
Institute,
Amsterdam.
Primigravidae
and multigravidae
with no history oftaking
antimalarial
drugs were studied
at the time oftheir first antenatal visit (day 0) and in subsequent follow-up visits at day 7 (primigravidae
only), day
28, and day 56. The exact location
of each woman’s
household
was determined
and mapped.
An additional
group of4l
women were screened at their first postnatal
attendances,
which were within 2-4 wk after delivery. As
controls,
15 nonpregnant
women (14 nulliparous),
were
screened on a single occasion. These women were attending
the hospital with minor complaints.
ET AL
MALARIA,
FOLACIN,
AND
HEMATOLOGY
Radioassay
Table I gives the normal
reference
values
for the radioassay.
Values for healthy pregnant
and nonpregnant
Dutch controls are given for
comparison
(18). Nonpregnant
RBC folacin
values are those of the West Kenya nuffiparous
control
group.
Samples were assayed in duplicate.
To assess
interassay
variation,
three control
pool sera
for cobalamin
and folacin were employed
on
all assay runs. Red cell hemolysates
were diluted (1:1) to approximate
the mean range of
the pool standards.
The standard
curve was
calculated
by spline function using a computer
TABLE 1
Radioassay
reference
Anemia,
folacin,
and cobalamin
status
Correlations
between
serum
and RBC folacin in primigravithe
and multigravidae
are
shown in Figures 1 and 2 for all blood samples.
Low serum folacin showed poor specificity
for
low RBC folacin concentrations
in this population.
High RBC folacin
values were observed
more
frequently
in primigravidae
(14.9%)
than
in multigravidae
(4.2%)
(p
< 0.0001).
Only two samples
from primigravidae compared
with nine from multigravidae
were found to have low RBC folacin
values
before delivery.
The mean serum and RBC
folacin concentrations
for multigravidae
were
significantly
lower
than
in pnmigravidae
(serum
folacin,
p < 0.03; RBC folacin,
p
< 0.001).
No difference
in mean cobalamin
concentrations
was observed
between
parity
groups. In the nonpregnant
control group, the
mean value for serum folacin was 9.0 nmol/l
(range 4.4-22),
for RBC folacin 655 nmol/l
(range 336-1139),
and for serum cobalamin
390 pmol/1 (range 220-750).
Tables
3 and 4 show the mean values of
hematological
parameters
at the first antenatal
visit, grouped
according
to gestational
age. The
highest mean values ofall parameters
occurred
at <12 wk gestation
for multigravidae.
Mean
Hb values for both primigravidae
and multigravidae
decreased
at 16 wk, but this fall is
not significant
and a further decrease in mean
Hb values did not occur later in gestation.
Primigravidae
had lower mean Hb values at all
stages of gestation.
The percentage
of women
values*
Source
Serum
folacin
nmol/1
Manufacturer’s
*
program
for smoothing
the curve (19). The
precision
of this radioassay
is shown
in
Table 2.
data
4.5-3l.7
RBC folacin
True cobslamin
pmol,4
nmol/I
283-1359
161-69lt
Nonpregnant
controls
Mean ± SD
4.4-22.0
(9.0 ± 3.5)
336-1139
(655 ± 223)
220-750
(398 ± 108)
Pregnant
controlj
Mean±SD
1.1-13.9
(4.8±2.2)
127-1457
(485±281)
187-664
(351± 108)
Range
of values.
t Indeterminate
cobalamin,
125-161
pmol/l; low cobalamin
< 125 pmol/l.
Gestation, 28 weeks. The cited range may indicate low folacin stores in a number
of these
women.
Downloaded from ajcn.nutrition.org by guest on June 11, 2014
mia had obvious
clinical malaria and she was
admitted
for treatment
and blood transfusion.
Mean ages (±SD) of primigravidae,
multigravidae,
nulliparous
controls,
and postnatal
groups were, respectively:
17.7 ± 2 yr, 24 ± 5
yr; 20 ± 3.8 yr, and 22. 1 ± 6.7 yr. All women
were from the Abasamia
subtribe and lived in
mud huts < 5 km from the hospital.
Study
samples
and control
groups
consisted
of
women
who were primarily
subsistence
cultivators.
While
economic
conditions
were
generally
poor, there was little evidence
of severe malnutrition
either in pregnant
women
or in children
under 5 yr (17). There was no
significant
difference
in the percentage
of primigravithe
or multigravidae
with ova of
hookworm
(62.9%
and 68.7%) or ascariasis
(23.6% and 22.4%), although
more primigravidae had trichunasis
(24.7%
and 13.4%, p
> 0.1). Ova counts
were not obtained.
A small
number were diagnosed
with other intestinal
infections:
S Mansoni
(six), E Histolytica
(fourteen),
G Lamblia
(four) and B Coli (one).
805
IN PREGNANCY
BRABIN
806
TABLE 2
Precision of radioassay
ET
AL
for folacin and cobalamin
Aay
Mean
Standard
deviation
Interassay
CV
Intraassay
variability
%
Serum folacin (nmol/l)
Pool 1
Pool 2
Poo13
Serum cobalamin
Pool 1
Pool2
Poo13
* Coefficient
4.23
6.88
11.14
0.29
0.39
0.80
Variability
duplicates
%
6.95 (l6)t
5.67 (17)
7.20(16)
±2.60(9)
34.0
2.6
13.5
±2.96
%
±1.4 (16)
±2.0 (16)
±1.0(15)
-
-
(pmol/l)
31.0
399.0
527.0
10.0
24.0
71.0
(16)
(17)
(16)
-
(8)
-
±1.0(15)
±1.6 (15)
±1.9(15)
of variation.
t Parentheses: number
of assay procedures.
above 300. Five of the postnatal
smear preparations were not ideal and for these, the segmentation
index was not determined.
None
of these five women
was anemic.
Folacin
The
women
concentrations
and parasitemia
pattern
of parasitemia
in pregnant
has been reported
previously
(20). In
summary,
peak
parasitemia
prevalence
cccurred at 13-16 wk gestation,
with primigravidae showing a higher peak prevalence
at that
time (85.7%) than multigravidae
(51.7%).
A
progressive
decrease
in parasitemia
prevalence
was observed
with advancing
gestation
such
that prevalence
figures were approximately
halved by late gestation
(primigravidae
42.9%;
multigravidae
23.5%).
The mean
RBC
folacin
concentrations
(±SD)
on day 0 for primigravithe
with parasitemia (1056 ± 400 nmol/l) was significantly
higher than in primigravithe
without
parasitemia (879 ± 374 nmol/l),
(p < 0.02, 1 17 degrees of freedom).
The difference
between
mean values for multigravidae
with parasitemia (879 ± 369 nmol/l)
and without parasitemia (832 ± 342 nmol/1) was not significant.
The mean parasite density for primigravidae
attending
on day
0 was significantly
higher
than that for multigravidae
(primigravidae
geometric
mean,
2100 parasites
per mm3;
multigravidae
398 parasites
per mm3;
p
< 0.001).
No correlation
was observed
between
folacin
concentrations
and parasite
density
in individuals.
As all 15 nonpregnant
controls had negative
Downloaded from ajcn.nutrition.org by guest on June 11, 2014
with anemia
(Hb < 1 1.0 g%) or with low folacin or indeterminate
cobalamin
values is
shown
in Table 5. An increase
in the percentage
of women
with anemia
occurred
at
13-26 wk gestation
but this difference
is not
significant.
Only one woman
had an Hb <6
g%. In primigravidae,
7.8% and 31.9% had Hb
values in the range of 6-8 g% and 9-10 g%,
respectively.
The corresponding
percentages
for multigravidae
were lower, 1.8% and 25.7%,
but not significantly
different.
Low RBC folacin
values
occurred
significantly more often in multigravidae
during the
third trimester
(p < 0.01), although
a similar
percentage
ofwomen
from both parity groups
had low values postnatally
(about 15%). The
percentage
of folacin-deficient
primigravidae
increased
only in the postnatal
period,
at
which time there was a corresponding
decrease
in the percentage
with anemia (p > 0. 1). The
percentage
of multigravidae
with anemia also
decreased
postnatally
(p = 0.06).
In individuals,
there was no correlation
between their Hb and RBC folacin values. Nonanemic
women
had a higher, but not significantly different, percentage
oflow RBC folacin
(5.9%) and indeterminate
cobalamin
(6.7%)
values than anemic
women
(2.0% and 2.9%).
Seven women had low RBC and serum folacin
values but only one of these was anemic
(Hb
10.4 g%). Four of25 smears from women with
low RBC folacin or indeterminate
cobalamin
values showed
anisocytosis
and two showed
hypochromia
with target cells. The range of
values for the segmentation
index was 248308 with one postnatal
smear with a value
FOLACIN,
MALARIA,
AND
HEMATOLOGY
2000
IN
PREGNANCY
807
I
I
PRIMIGRAVIDAE
I
-
I
1800
.
I
I
I
I
I
II
I
II
1600
I
I
I
#{149}
I
I
I
#{149}
I
1
1400
I
I
II
I.
I
I
I.
I
I
I
I
1200
I
#{149}
I
#{149}
I
I
#{149}
#{149}
#{149}
#{149}
I
#{149} I
I
#{149}
#{149}
I
S.
#{149}
Downloaded from ajcn.nutrition.org by guest on June 11, 2014
II
I
I
I
#{149}
#{149}#{149}#{149}
#{149}I
I
1000
#{149}
II
#{149}
#{149}
RED CELL
FOLACIN
N.MoI L
#{149}.
#{149}
#{149}
#{149}
800
#{149}.
I
#{149}III
I
#{149}I
I
S.
II
#{149}4#{149}#{149}
I
I
#{149}
#{149}I
#{149}I
I
I
#{149}
I
I$III
I
#{149}
II#{149}
I#{149}I
#{149}II
I
II
I
I#{149}
1111
I
I
II
11111
I
II
I
#{149}
600
t
#{149}II
I
II
III
I
$
I
II
#{149}#{149}#{149}
II
II
II
#{149}
#{149}#{149}#{149}:#{149}
.#{149}
II#{149}
I
I
#{149}I
I
I
I#{149}
I
II.
I
I,#{149}
II
400
I
I
I
I
I
-
200
LI
1
2
3
44.55
6
SERUM
+
7
FOLACIN
Mid-point
8
9
10
11
13
14
15
N.MoI.L.
mean
valueB
FIG 1. Correlation
between serum and RBC folacin in primigravidae.
High and low indicate
distribution
for nonpregnant
adults. + indicates
midpoint
mean values.
malaria
smears,
the mean value of their hematological
parameters
have been compared
with those
of pregnant
women
without
parasitemia
on day 0 (Table 6). Primigravidae
12
the limits of the normal
have significantly
higher mean RBC folacin
and significantly
lower Hb than controls
(p
<0.05),
and multigravidae
significantly
lower
mean Hb and Hct than controls
(p <0.005).
808
Fr
BRABIN
AL
2000
I
MULT
IGRAVIDAE
I
1800
I
I
I
1600
HIGHI?.
#{149}
I
I
1359
I
I
#{149}
I
1200
I
Downloaded from ajcn.nutrition.org by guest on June 11, 2014
I
II
I
I
I
I
II
I
I
1000
#{149}I
I
#{149}1
REDCELL
FOLACIN
N.MoI.L.
$
#{149}
I
#{149}
II
I#{149}.I
I
1#{149}1
#{149}
j.I
%
II
::
I
I
II
I
11
III
I
#{149}
I
#{149}
#{149}.
#{149}
I
I
#{149}#{149}
#{149}1..
III...
a.
#{149}
III.
#{149}.
I#{149}
I
I
I
I
#{149}
II
.;.,
400
#{149}
I#{149}
#{149} #{149}
I.
I
#{149}
#{149}
1:1511::
.
.1
:
I
3
I
I
.I#{149}S.I#{149}#{149}
600
#{149}
I.
Ill
I
:.#{231}#{149}1
#{149}
I
800
I
I#{149}
.
#{149}
#{149}
I
LOW
283
.
200
I
LOWI
i-4
r
1
2
3
44!55
6
SERUM
+
FIG 2. Correlation
distribution
Red
between
for nonpregnant
cellfolacin
serum
adults.
in follow-up
Mid-point
midpoint
and
age in pri-
serum
‘
‘
‘
9
10
11
mean
mean
12
13
14
15
N.MoI.L.
in multigravidae.
periods
Tables 7 and 8 show the mean
RBC folacin
values by gestational
‘
8
FOLACIN
and RBC folacin
+ indicates
.
7
values
High and low indicate
the limits of the normal
values.
migravidae
and multigravidae
at day 0, day
28, and day 56. Classification
is according
to
the gestational
age when the patient was first
seen and the vertical
columns
remain
true to
MALARIA,
I
I
FOLACIN,
I
I
AND
HEMATOLOGY
IN PREGNANCY
I
r-
00
+1
+I’
-
-
r
e
+1on’
000
-
,
-
#{226}
Iiri
on
-
ret
a
+i’TO
,
IFt
a
+1
00
a
0
-
+1
+1
+1
+10
+1
00
O
.
re
+1-
-
----‘
m
-
---.-
809
00
r-
ret
.ret
-
IF-I
o
0
.
-
.
-
ret
41
r
-H7
+I
0r-’
+l
---o--
+I
+1.
+I
41
+I
ret ,
ret
-
+‘-r
+I
41
+1
00
‘
00
-
.
ret
-
.
ret
-
00
-
ret
6
C
:
IF)
+1-IF-,
4-Irt
-H’
Ho
+1’r
I
-:
o
41
+1
o
00
.
00
q00
I
I
©
#{247}
+I
N
+IZ’
H’
.
N
H
r1
.1I
H
ret
IFt
+1
IF
I
I
+I’7
+l
IFt
O
‘
+1
I
-
ret
IF
ret
+1
I
+I
1’
IF
-
+1
-
41Z
:
N
+1
I
0
*
+-
+1
-r
.U
-
ii
+ITo
+Ioo
ri
‘,a
fh0
+Ir
66
+1
(-I
w
IF-,
eN
0
-
+1
rn
on
r-i
66
IFi
0
-
‘.5
60
I0t
I
I
60
cn
I
5.
E
V
kbh
I_c
6
V
+1
6
oo
EEEE”E
*
+-*+C.O3
I
I
c’i
.
60
.;
+1
!
EEEE”
*
+.-+4.cO
Downloaded from ajcn.nutrition.org by guest on June 11, 2014
sr..
+1
BRABIN
810
TABLES
Prevalence
(%) of anemia,
low folacin,
and indeterminate
Anemia
Group
22.2
40.0
44.8
0.0
24.0
37.5
Postnatal
21.1
9.5
13.3
forfolacin
by gestation
PG
MG
28.6
30.0
24.1
-
26.7
andcobalamin(manufacturer’sdata,
PG
MG
0.0
3.7
15.6
15.8
l4.3
0.0
Table
0.0
6.1
6.9
<1 1 g% (World
0.0
6.1
0.0
-
-
0.0
-
1); forHb
MG
125-i6Jpmol/I
0.0
1.5
0.0
-
visit
Indetenmnate
cob.lamin
<283 nmoi/l
25.0
31.2
31.2
-
(wk) on first antenatal
Low RDC folacin
<4.Snmoi/I
Pregnant
12 wk
13-26wk
27-40 wk
Nulliparous
a Cut-offpoints
values
folacin
PG
<Jig
AL
cobalamin
Low serum
MO
POt
ET
-
Oipnization).
Health
t PG indicates primigravidae
and MG, multigravidae.
t Postnatal samples available only for RBC folacin.
§ Refers to nonpregnant
control group.
tational
stages
is significant
for multigravidae
(RBC
folacin,
p < 0.01;
serum
folacin
p
< 0.001)
(Friedman
test; nonparametric
analysis of variance)
and for primigravidae
(RBC
folacin
p < 0.01; serum folacin p < 0.025).
These
changes
were not seen in the crosssectional
study ofthis population
attending
at
day 0 (Tables 3 and 4), nor were they observed
in cross-sectional
analysis ofthe mean values
for women attending
at day 28 or day 56. The
decrease
in mean folacin
values
observed
in
Tables 7 and 8 cannot conclusively
be attrib-
TABLE 6
Hematological
parameters
ofpregnant
and nonpregnant
Discussion
Radioassay
A good correlation
is expected
between
microbiological
and radioassay
methods
with
large numbers
of samples
(14, 2 1). A highly
significant
correlation
for cobalamin
values
can be obtained
in samples
from pregnant
women measured
by the two methods (22, 23),
women
Pregnan
Parameter
I
<
0.05;
1p
< 0.005.
Significance
MUItigraVidse
6.9 ± 3.0 (20)1
900 ± 417 (2l)t
323 ± 106(17)
11.7 ± 1.7 (21)t
37.6 ± 4.8 (21)
of women.
levels against
parasitemia
values
seen on day 0.
women
Primigravidae
Serum folacin (nmol/l)
RBC folacin (nmol/l)
Serum oobalamin (pmol/l)
Hb(g%)
Hct (%)
Mean ± SD; parentheses:
number
t p
without
t
5.5
809
360
11.7
34.7
±
±
±
±
±
for nonpregnant
2.2
349
125
1.6
5.5
(58)
(59)
(56)
(59)f
(59)
controls.
Nonpregnant
6.6
655
454
13.1
39.9
±
±
±
±
±
women
2.6 (10)
223 (15)
336(10)
1.5
(15)
3.3
(15)
Downloaded from ajcn.nutrition.org by guest on June 11, 2014
uted to gestational
changes alone, since these
changes
were not observed
in the crosssectional
analysis.
An important
difference
between
women
seen at different
follow-up
times, apart from gestational
age, is the administration
ofchloroquine
at days 0 and 28.
Serum
cobalamin
concentrations
showed
no
significant
decrease
in mean values during the
follow-up
period.
the same
individuals.
Measurements
from
women
who made only a single clinic visit
have not been included
at day 0.
In both parity groups, a progressive
fall in
mean values occurred
between
day 0 and day
28, and between
day 28 and day 56 for most
gestational
stages.
This decrease
in mean
serum and RBC folacin values at different ges-
MALARIA,
FOLACIN,
AND
HEMATOLOGY
IN
PREGNANCY
811
ri
II
II
II
N
cu
.
,.
O
o
o
-‘
N
V
-
;
E
.
N
IF-
.
N
0
ret oo
-
N
.
a’
ret
N
N
V
#{149}E
-0-onm
g
g
0
00
-
00
o\
‘!t-%r.4
ret
+I2+12+I+1+I
oo___.
0
L)
00
_
ret
N
N
O
N
r
__.,It.__.
-_-_
+1+l2+l2+1
#{226}
-___IF-,
c
_
_
6
1f
0
0
se
V
8
§
Nret4
:
_
.
_oo
e4
a ‘‘c
#{149}
:
If
a
#{149}
8
‘‘rt
.
ret N
+i+I’ret
.
N
N
N
‘
ret
+IN‘N+IN
N
‘
N
“-V
ret
:,,j:
N
_
N
N
-
ao
ret
N
N
reon
0N
cu
.
2
C
I
#{149}0
E
I
.
iiis’isiRi1
N
a
-oo
-
“3.
-
Z?
‘
‘...-N
5
tu
5
6
-
C
6
..,a
.6
cu
cu
.6
u,
I
*
+i+i:+i+ig:+i:+i6
.
It
IF
0ret
::
IF
N
If)
6
-
.c
J,,,
:t
.
c
.
IF)
.6
-
.
a
N
N
O
N
If)
N
,ri
,
.6
V
.E
.-
VE
*4.
‘ft
.
N
0
Vt
,
#{247}I+1Z
N-
+1+1
am
+lN+I
H
NIFNIF)N
c
.o
r
r
=
9
u
g
*
00
N
.0
If)
!
,.
.0
once c
rl)
._-,-_______-_--._-___,
.-.------,
0
-
6
IL.
rI)
N
:
I+I
5
N
e
66
Ec.
e
+I
n
fi
6
V
)
0
c
.
6
<.
.
n
‘0
3
F-
I
3IF)
I1
V
N(
8
c
8
re;
N
r
.
:.
.6
8
.
.
n
6
.
0.
ret
.
.
‘a
0
.-
+-4+
-
8
8
I),
&)
L)
-
c/)
I
Downloaded from ajcn.nutrition.org by guest on June 11, 2014
,
.
BRABIN
812
although lower values with the microbiological
method
have been reported
using pregnancy
sera (22). Both RBC and serum folacin, at all
stages of pregnancy,
show a broad variance
and a low correlation
between
paired samples
(Figs 1 and 2). Other studies in pregnancy
utilizing radioassay
have reported a similar weak
correlation
between
for a healthy
women
(18).
Gestational
hematological
pregnant
serum
and
RBC
population
values
these male workers, and it must be considered
that their low mean Hb values are a result of
both malaria and iron deficiency.
Surprisingly,
only one woman
had severe anemia
(Hb < 6
g%), which suggests
that the combined
effect
of helminthic
and plasmodial
infection
in a
highly endemic
rural area is to produce
moderate but rarely severe anemia in pregnancy.
Unexpectedly
and parity
values
differences
in
anemia
in
primigravidae
is
high RBCfolacin
values
Relation
to Pfakiparum
infection.
There is
evidence
in the present study (Figs 1 and 2)
that RBC folacin
concentrations
may be cxceptionally
high in pregnant
women
living
under holoendemic
conditions
for malaria.
The normal range of values for the nonpregnant control
group
(336-1 139 nmol/l)
is
comparable
to that for other nonpregnant
reference
groups when the same radioassay
procedure
is used (Table
1). Serum folacin
values were not higher than the accepted
normal range. Quality
control
was satisfactory
throughout
the assay procedure
(Table 2), and
meticulous
care was taken with the handling
and storing ofhemolysate
samples in the field.
The significant
decrease
in folacin values in
the longitudinal
groups of primigravidae
and
multigravidae
following
chioroquine
ingestion
(Tables 7 and 8) provides further evidence that
the raised values relate to malarial
infection.
However,
proof that malaria produces
raised
RBC folacin concentrations
remains to be established
and, at present, the association
cannot be considered
cause and effect on the basis
of the present study.
One likely explanation
of these changes
in
RBC concentrations
would be an increase
in
the young red cell population
in individuals
who are actively
hemolyzing.
This is because
reticulocytes
have a much higher RBC folacin
concentration
than mature red cells (31-33).
This would also explain the raised RBC values
in women without parasitemia
on day 0 (Table
6), because
ongoing reticulocytosis
may occur
after the suppression
of parasitemia
(34). If
this hypothesis
is correct,
then RBC folacin
measurement
would
be a poor measure
of nutritional
folacin
status and usable only if a
method
of correcting
for the degree of reticulocytosis
were developed.
However,
there are reasons for considering
that raised RBC folacin values may also result
Downloaded from ajcn.nutrition.org by guest on June 11, 2014
of
to be higher in women
from endemic
malarial
regions (20, 27, 28), and this has been
significantly
associated
with their higher preyalence
of malaria
(29). Hemolysis
would be
expected
to decrease
after delivery
when malana prevalence
falls (20), and this may partly
explain
the lower percentage
of women
with
postnatal
anemia
(Table 5).
Despite
the prevalence
of parasitemia,
the
women
in this study
generally
remained
asymptomatic.
In such a situation,
a low-grade
hemolytic
anemia
could be wrongly
considered primarily
nutritional
or secondary
to helminthic
infection.
The mean levels of Hb in
the study groups are much lower than values
reported for male road workers located in Kisumu on the Lake Victoria
shoreline
(30). Kisumu is a large town and also a holoendemic
malarial
area, but 70 km south of the present
study location.
Women
in the present study
had a higher percentage
of infections
with
hookworm,
ascariasis,
and Pfakiparum
than
known
AL
of Dutch
There is a decrease
in mean Hb at 12-20
wk gestation
and an earlier fall in mean
Hct
at 12- 16 wk gestation
(Tables 3 and 4). These
early differences
in Hb and Hct are likely related to plasma volume changes (24), although
no further fall in mean values occurs with advancing
gestation,
as others report (24, 25). A
rise in the percentage
of women
with anemia
in the second trimester has also been reported
in a malaria-endemic
area (26) and may relate
to hemolysis
as it follows the pregnancy
period
of peak malaria prevalence
(13-16
wk gestation). Hemolysis
would also explain why primigravidae,
who have the higher prevalence
of P fakiparum
infection,
also have lower
mean Hb values throughout
pregnancy.
The
prevalence
ET
MALARIA,
FOLACIN,
AND
HEMATOLOGY
813
PREGNANCY
were present in this series. What then could
be the basis for any nutritional
superiority
of
these rural Kenyans?
First, there are difficulties
with all earlier
studies: none report RBC folacin values in relation to malaria infection,
so there is no direct
evidence
that the megaloblastosis
that occurs
is due to folacin deficiency.
There is evidence
in children
that dyserythropoietic
changes
in
severe anemia
due to Pfakiparum
are not a
consequence
of cobalamin
or folacin
deficiency within marrow cells (9). Serum folacin
values alone are inadequate
to substantiate
the
relation
of malaria
to megaloblastosis
in view
of their low specificity
for low RBC folacin
concentrations.
Second, recent studies in children indicate that RBC folacin values may be
raised in association
with malaria.
Higher folacin values are reported
in convulsing
chilthen from Nigeria
compared
to a nonconvulsing group (47), and more ofthe group with
seizures had malaria. In Gambia,
children
with
malaria had high RBC folacin values measured
by microbiological
assay (8) as well as by the
Becton-Dickinson
Radioassay.
In the present
study, umbilical
cord venous
blood samples
were available
from a small number of infants
of women
screened
in pregnancy
who later
delivered
at the hospital.
These did not demonstrate
low cord plasma or RBC folacin activities and some samples
showed
high RBC
folacin values, although
this could not be related to Pfakiparum
placental
infection
(48).
Folacin
metabolism
in mature
red cells is
relatively
inactive.
However,
if high intraerythrocytic
folacin
concentrations
can be
produced
in immature
cells through
a biochemical
pathway
involving
the malaria parasite, then mean RBC folacin values would
take a number of months to return to normal
after the suppression
of parasitemia.
As malana is a chronic
infection
in immune
pregnant women
living under holoendemic
conditions,
the occurrence
of high intraerythrocytic folacin
concentrations
could
influence
nutritional
when dense
the placental
status,
particularly
parasitemias
villi.
may
in pregnancy
be present
in
We are grateful to the Provincial
Medical
Officer in
Kakamega,
and the District Medical Officer in Busia, for
permission
to work in this area ofKenya.
Thanks are due
to Dr DJB Wijers and Dr TAC Hanegraafforlogistic
sup-
Downloaded from ajcn.nutrition.org by guest on June 11, 2014
from factors additional
to reticulocytosis.
In
experimental
animal
systems,
raised folacin
values have been reported (35-37),
and these
changes
have been correlated
with the intraerythrocytic
development
of the parasite
in
vivo (38). Siddiqui
and Trager (36) consider
that these increased
concentrations
are produced from an altered metabolism
ofthe host
red cell itself, rather than from folacin synthesis by the parasite.
Further
evidence
for
their hypothesis
is that folic acid reductase
inhibitors
are effective
against the parasite only
as long as it remains
inside the erythrocyte
(39). Inasmuch
as immature
red cells are preferentially
invaded
by malaria
parasites
(40),
high concentrations
of folacin may preferentially occur within these cells when infected.
This would complicate
any simple
method
developed
to correct for the raised folacin values normally
present in reticulocytes.
A biochemical
mechanism
producing
high RBC folacin concentrations
may also explain the lack
of effect of folic acid supplements
on parasite
density in pregnant
women
(41) and in children (42).
Relation
to megaloblastosis.
In the study
population,
megaloblastic
anemia
is possibly
present
in only a small number
of women.
Evidence
for this is based on the observation
that neutrophil
hypersegmentation
was not
observed
in any blood smears from women
with low levels of RBC folacin
or indeterminate cobalamin,
but it is not conclusive
because marrow
aspirates
were not available
for
diagnosis,
and it is established
that megaloblastic erythropoiesis
may be present in the
absence
of hypersegmented
neutrophils
and
anemia
(43). Nevertheless,
the finding
was
unexpected
and
might
be considered
exceptional in view of the observations
of Fleming
et al (44, 45) who reported
a high incidence
of megaloblastic
anemia
in pregnancy
in Nigeria that was greatly reduced
with prophylactic
pyrimethamine.
A similar result from
antimalarial
prophylaxis
has been
shown
by
Lawson (27) and among selected groups of urban hospital admissions
in Nigeria and Kenya,
a high proportion
ofprimigravidae
have been
reported
with megaloblastic
anemia
(4, 5).
Folacin
deficiency
complicating
malaria
has
been described
in adult men in Vietnam
(46);
low serum folacin and megaloblastic
marrows
IN
814
BRABIN
port from the Medical Research
Centre, a department
of
the Kenya Medical Research Institute,
Nairobi; to Nanghia
Hospital;
to MA Baatenburg
de Jong-Schouten
ofthe Department
ofHematology,
Academic
Medical Centre, Amsterdam;
and to Dr I. Chanarin
for his comments
on an
early draft ofthis paper. This paper is published
with permission ofthe Director
of Medical Services for Kenya.
References
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
AL
ration with R proteins blocked by vitamin B12 annlogues. J Clin Pathol l982;35: 1110-3.
Brabin LR, Brabin BJ, Liboon L Referral of children
for nutritional
intervention
in an under-fives
clinic:
would weight for height assessment
help? A J Trop
Med Hyg l984;87:9l-7.
van den Berg H, Bruinse HW. On the role of nutrition
in normal human pregnancy.
University
of Utrecht,
1983. PhD Thesis.
Marscher
I. Calculation
of the radioimmunoassay
standard
curve by spline function.
In: Radioimmunoassay and related procedures
in medicine.
International
Atomic Energy Authority,
Istanbul meeting;
1973:111-22.
Brabin 8.1. An analysis
of malaria in pregnancy
in
Africa. Bull WHO I983;61:1005-l6.
Newmark
PA, Gordon
YB. Radio-immunoassay
and
saturation
analysis.
Br Med Bull l974;30:86-9.
Green R, Colman
N, Metz J. Comparison
of results
of microbiologic
and radio-isotope
assays for serum
B,2 during pregnancy.
Am J Obstet Gynecol
l975;122:
21-4.
Kalameghan
R, Krisnaswamy
K. Functional
significance oflow serum B12 levels in pregnancy.
Int J Vit
Nutr Res l977;47:52-6.
Prema K, Ramalakshmi
R, Madhavapeddy
R, et al.
Changesin
haemoglobin
levels duringdifferent
periods
ofgestation.
Nutr Rep Int 198 l,23:629-36.
National
Research Council.
Laboratory
indices of
nutritional
status in pregnancy.
Washington
DC: National Academy ofSciences,
1978.
Rougemont
Par A, Boisson ME, Dompnier
JP, et al.
Paludisme
et anemic de la grossesse en zone de savane
africaine.
Bull Soc Pathol Exot l977;70:265-73.
Lawson
JB. Malaria
in pregnancy.
In: Lawson
JB,
Stewart DB, cdi. Obstetrics
and gynaccology
in the
tropics and developing
countries.
London:
Edward
Arnold,
1967:59-72.
Kortmann
H. Malaria and pregnancy.
University
of
Amsterdam,
1972. MD Thesis.
van Dongen
PWJ, van’t Hof MA. Sickle cell trait,
malaria and anaemia
in pregnant Zambian
women.
Trans Roy Soc Trop Med Hyg 1983;77:402-4.
Latham MC, Stephenson
LS, Hall A, Wolgemuth
JC,
Elliot TC, Crompton
DWT. A comparative
study of
the nutritional
status, parasitic
infections
and health
of male roadworkers
in four areas of Kenya. Trans
Roy Soc Trop Med Hyg l982;76:734-40.
Izak G, Rachmilewitz
M, Grossowitz
N, Galaveski
K, Kraus SH. Folate activity in reticulocytes
and the
incorporation
of tritiated pteroylgjutamic
acid into
red cells. Br J Haematol
1968;l4:447-52.
Aung-Than-Batu,
Hia-Pe U. Studies ofthe first Asian
congress of nutrition.
Hyderabad:
Nutrition
Society
oflndia, 1971:122-33.
Hofibrand AV, Newcombe
BFA, Mollin DL MethOd
ofassay
ofred cell folate activity and the value of the
assay as a test of folate deficiency.
J Clin Pathol
l966;l9: 17-28.
WOOdrUffAW,
Ansdell YE, Pettit LE. Cause of anaemm in malaria. Lancet 1979;l:l055-7.
TragerW. The enhanced folic and folinic acid contents
of erythrocytes
infected with malaria
parasites.
Exp
Parasitol
1959;8:265-73.
Downloaded from ajcn.nutrition.org by guest on June 11, 2014
1. Foy H, Kondi A. Megaloblastic
anaemia in East Africa. In: Tropical sprue and megaloblastic
anaemia.
Welcome
trust collaborative
study. Churchill,
1971:
26 1-7.
2. Adams EB, Hift W. Nutritional
and other megaloblastic anaemias among Africans and Indians in Durban. Trans Roy Soc Trop Med Hyg 196 1;55:374-82.
3. Lothe F, Patal KM, Tozer RA. MegaIObIaStIC anaemia
in Uganda.
Trans Roy Soc Trop Med Hyg 1969;63:
393-7.
4. Mati JKG, Hatimy A, Gebbie DAM. The importance
of anemia of pregnancy
in Nairobi and the role of
malaria in the etiology of megaloblastic
anemia. Am
J Trop Med Hyg 197 l;74:1-8.
5. Ojo OA. The pattern ofanemia
in Western
Nigeria.
Am J Trop Med Hyg l965;68:32-6.
6. Edelstein T, Stevens K, Brandt V. Baumsiag
N, Metz
J. Tests of folate and vitamin
B12 nutrition
during
pregnancy
and the puerperium
in a population
subsisting on a sub-optimal
diet. Br J Obstet Gynaecol
l966;73: 197-204.
7. Fleming AF, Stenhouse
NS. Folate activity, vitamin
B12 concentration,
and megaloblastic
erythropoiesis
in anemic
pregnant
Nigerians.
Am J Clin Nutr
l969;22:755-66.
8. Abdalla
5, Weatherall
DJ, Wrickramasinghe
SN,
Hughes
M. Anaemia
of P. falciparum
malaria. Br J
Haematol
1980;46: 17 1-83.
9. Abdalla
SH, Wrickramasinghe
SN, Weatherall
DJ.
The deoxyuridine
suppression
test in severe anaemia
following Pkismodiumfalcipanun
malaria. Trans Roy
Soc Trop Med Hyg 1984;78:60-3.
10. Brabin BJ. Hypothesis
the importance
of folacin in
influencing
susceptibility
to malaria infection
in infants. Am J Clin Nutr 1982;35:146-5l.
1 1. GhitiS J, Lora C. The folate binding in milk. Am J
Clin Nutr l967,20:l-4.
12. Metz J, Zalusky R, Herbert
V. Folic acid binding by
serum and milk. Am J Clin Nutr l968;21:289-97.
13. Costa M, RothenbergSD.
Appearance
offolate binders
in leukocytes
and serum ofwomen
who are pregnant
or taking oral contraceptives.
J Lab Clin Med 1974;83:
207-14.
14. Dunn RT, Foster LB. Radioassay
ofserum
folate. CIin
Chem Acts 1973;19:l 101-5.
15. Kolhouse
JF, Kondo H, Allen NC, Podell E, Allen
RH. Cobalamin
analogues
are present
in human
plasma and can mask cobalamin
deficiency
because
current radio-isotope
dilution
assays are not specific
for true cobalamin.
N Engl J Med l978299:785-92.
16. 8am R, Broom
GN, WOOdside
J, LitWiniZUk
RA,
Wrickramasinghe
SN. Assessment
of a radioisotope
assay for vitamin
B12 using an intrinsic factor prepa-
ET
MALARIA,
36. Siddiqui
FOLACIN,
WA, Trager W. Comparative
offolic and folinic acids oferythrocytes
37.
38.
39.
40.
41.
42.
AND
HEMATOLOGY
bioautography
and livers of
parasites.
J Par-
PREGNANCY
815
43.
Chanarin I, Rothman
D, Ardeman 5, Berty V. Some
observations
on the changes preceding
the development of megaloblastic
anaemia
in pregnancy
with
particular
reference
to neutrophil
leucocytes.
Br J
Haematol
l965;l 1:557-62.
44. FIeming AF. Haematological
manifestations
of malana and other parasitic
diseases.
Clin Haematol
198l;l0:983-lOl 1.
45. Fleming AF, de Hendrickse
JP, Allan NC. The prevention ofmegaloblastic
anaemia
in pregnancy
in Nigeria. Br J Obstet Gynaecol
1968;75:425-32.
46. Strickland
GT, Kostinas
JE. Folic acid deficiency
complicating
malaria. Am J Trop Med Hyg 1970;19:
910-5.
47.
Osifo BLA, Lukanmbi
FA, Fami1USi JB. Comparison
of folate levels in convulsing
and non-convulsing
febrile children.
Trop Geog Med l983;35:363-8.
48. Brabin BJ. A comparison
ofmaternal
and fetal folacin
and cobalamin
activities at parturition in relation to
P. fakiparum
infection.
Trans Roy Soc Trop Med
Hyg l985;79:652-6.
Downloaded from ajcn.nutrition.org by guest on June 11, 2014
normal ducks infected with malaria
asitol l964;50:753-6.
Reid YE, Friedkin M. Plasmodium berghei: folic acid
levels in mouse erythrocytes.
Exp Parasitol 1973;33:
424-8.
Siddiqui
WA, Trager W. Folic and folinic acids in
relation to the development
ofPlasmodium
lophurae.
J Parasitol l966;52:556-8.
Peters W. Chemotherapy
and drug resistance
in malana. New York: Academic
Press, 1970:41.
Pasvol G, Weatherall
DJ, Wilson RiM. The increased
susceptibility
of young red cells to invasion
by the
malarial
parasite.
Br J Haematol
l980;45:285-95.
Gail K, Herms V. Der einflub von pteroylglutaminsaure(Tolsaure)aufdie
parasitendichte(P/asmodium
falciparum)
bei schwangeren
frauen in West Africa.
Zeit Tropenmed
Parasitol
1969;20:440-50.
Huriey MGD. Administration
ofpyrimethamine
with
folic and folinic acids in human malaria. Trans Roy
Soc Trop Med Hyg 1959;53:410-I.
IN
`