Falciparum malaria and pregnancy

427 TRANSACTIONS

OPTHEROYAL

OPTROPICAL MEDICINE

SOCIEIY

Falciparum

AND HYGIENE,

VOL.

73, No.

4, 1979

malaria and pregnancy

R. S. BRAYSAND M. J. ANDERSON~ MRC Laboratories, Fajara, The Gambia

Summary

During pregnancy an increase occurs in the prevalence and density of Plasmodium falciparum malaria. The increase relative to non-pregnant women remains fairly constant with age though there is an over-all decrease in prevalence and density in both groups with age. The increase is at a height relatively early in pregnancy and declines after mid-term. At the height of the malaria season in The Gambia less antimalarial antibody appears in cord bloods than in the early dry season when transmission is lower. Infants show an increase in malaria prevalence and density in the early dry seasonas compared to the mid-wet seasonwhile the remainder of the population shows a decrease. Introduction

Malaria in The Gambia is seasonal and most transmission occurs in the months of July to November. If infants in the nought to six-months age group are protected by passively acquired maternal anti-malarial IgG, then the degree of protection conferred may vary seasonally since maternal antibody titres will be lower in the dry non-transmission season. This may put the infant at a particular disadvantage in the ensuing wet season and this in turn might be reflected in the seasonal variations in infant parasite rates and densities. Furthermore most reports of antimalarial IgG in cord and mothers’ blood in West Africa show that levels are equal or lower in cord blood (SEITZ, 1971; WILLIAMS & MCFARLANE, 1970; MCGREGOR& WILSON, 1971) whereas normally IgG levels are slightly higher in cord blood than maternal blood (though not necessarily so in rural West Africa). The auestion arises whether maternal malaria causes a reduction in the passageof specific antimalarial IgG across the placenta. It is generally agreed that during pregnancy women show an increased prevalence and density of malaria (BRUCE-CHWATT,1952; MADECKI & KRETSCHMAR, 1966; KORTMANN,1972), but it is not known if this increase relative to non-pregnant women remains unaltered with parity and/or age. There is, rather surprisingly, evidence that the greatest increase in infection occurs in the second trimester (PINGOUD, 1969), and that the anaemia consequent upon increased prevalence and density in primiparous women occurs between 16 and 24 weeks of pregnancy (GILLES et al., 1969) but the evidence requires confirmation. We have opposing hypotheses on the role of the placenta in malaria: one, that increased malaria is due to the presence of a placenta-a prime and possibly privileged site for the development of the older forms of P. falciparum

(CLARK, 1915)-the other, that the presumed recrudescences of falciparum malaria following parturition are due to the loss of the highly developed cellular immune system contained in the placenta which has been acting as a brake on malaria (GARNHAM, 1938). Finally, is there an increased number of recrudescences of falciparum malaria during the period immediately after delivery, as has often been assumed? An attempt has been made to shed light on these points, firstly by a blood film and malarial antibody survey of pregnant women and infants of nought to five months at four periods in the year and a malarial antibody survey of cord bloods during the year of 1976 and secondly by another survey of pregnant and non-pregnant women in the wet season of 1977. Materials

and Methods

In February, May, August and November 1976 surveys were conducted in villages in the Lower River Division of The Gambia, including Sanquia, Kaiaf, Genieri, Massembe, Kwinella, Madine Angaly, Sibite, all villages near to the-provincial caoital Mansakonko and at Mansakonko Health Centre’s ante-natal clinics. Children less than six months old and pregnant women were examined by taking finger prick blood both on slides for staining and in Carroway tubes for serum separation and collection. The age of the child and the week of the pregnancy were noted. Cord blood was collected and serum separated from placentae of mothers delivering at Mansakonko Health Centre and in the village of Keneba in the Lower River Division. Dr. Joan Bryan kindly made collections of Anopheles resting in huts in the villages concerned to check for sporozoites in May, August and November 1976 and in February 1977. A total of 329 pregnant women and 404 infants was examined. 52 cord bloods were collected. During the 1977 malaria season, i.e. August, Seotember and the first half of October. blood films weie collected from 1,000 pregnant and 1,000 nonpregnant women at ante-natal and post-natal clinics at the Privincial Health Centres of the Western Division (WD), Lower River Division (LRD) and McCarthy Island Division (MID). Numbers were matched at each clinic and details were collected concerning gestational dates and parity among i Present address: Imperial College Field Station, Ashurst Lodge, Sunninghill, Ascot, Berks, UK 2 Present address: Department of Internal Medicine San Diego Naval Regional Medical Center, San Diego, California, USA

428

Falciparum

MALARIA

pregnant women and the age of the last child and parity among non-pregnant women. Serum was collected from 300 pregnant and 300 non-pregnant women at the Mansakonko clinic, LRD. Any women given antimalarials at the clinic (a very low proportion) were omitted from the survey. Thick films were stained with Giemsa and 100 high power microscopic fields examined. Parasites per 500 w.b.c. were counted and the number multiplied by 10 and expressed as parasites per mm3 as another survey had shown the normal w.b.c. count of Gambian women to be close to 5,000/mm3. The overwhelming majority of infections were due to P. falciparum and only this parasite is considered here. Serum was separated and tested in the immunofluorescent test (IFT) using cultured P. falciparum schizonts as antigen. Table I-P. falciparum density and antibody

malaria

in infants

AND PREGNANCY

Results

Mosquito surveys confirmed that sporozoites were found in the salivary glands of Anopheles in the August and November surveys only. It was difficult to find Anopheks in houses in February and impossible in May. A survey of 100 pregnant and 100 non-pregnant women showed no significant difference in their white blood cell count. Table I shows the results of the parasite and antibody surveys of pregnant women and infants and the antibody levels in cord blood. It can be seen that the pregnant women showed a rise in prevalence and density of parasitaemia and specific antibody levels in the wet season month of August; these levels then dropped by the early dry season month of November, just like other adults and children (unpublished observations). This was not mirrored in the specific antibody levels in the cord

and pregnant

women:

seasonal

variations

in prevalence,

February (Dry)

May (Dry)

ihlgust (Wet)

12.5% 51 1.10

15% 194 0.74

11% 422 0.99

29% 1391 0.98

Prevalence Mean parasite/mm3 Mean log,, IFT titre

3% 79 1.25

07p 0.91

15% 532 0.81

32% 2232 0.98

Pregnant women Prevalence Mean parasite/mm3 Mean log,, IFT titre

34 % 287 2.38

28 % 244 1.95

40% 1972 3.07

33% 524 2.18

2.17

1.82

1.58

2.18

Mean density log (n + 1)

2 x S.E.

1.59 O-98 0.62 0.74 0.53 0.92

0.19 0.21 0.16 0.20 0.11 0.08

0.81 0.61 0.47 0.41 0.38 0.58

0.12 0.15 0.13 0.13 O-11 0.06

November (Early dry)

Infants O-5 months

Prevalence Mean parasite/mm3 Mean log,, IFT titre Infants 2-3 months

Cord blood

Mean log r,, IFT titre

Table

II-Prevalence

and density

Category

Total

Primio Para i Para 2 Para 3

255 171 189 123 262 1000

pT”a:“a? 4 Para 1 Para 2 Para 3 Para 4 pT”o:“,? 5

320 176 172 142 190 1000

of P. falciparum

Infected

malaria

in women

% Prev.

Malaria in Pregnant Group 151 59.1 75 43.9 59 31.2 :: 413

38.2 30.9 41.3

Malaria in Non-pregnant Group 146 45.6 65 36.9 ii

31.4 29.6 24.2 35.3

429

R. S. BRAY AND M. J. ANDERSON

Table

III-Prevalence

Weeks Pregnant Q 24 25-32 33-40

and density

of P. falciparum

malaria

in pregnant

women-Period

Mean Total

Infected

276 450 274

139 176 98

Table IV-Prevalence and density of P. falciparum malaria in pregnant women: ratio of pregnant/non-pregnant by parity Mean density Category

Prevalence

log (n+l>

E”3 px PsP4 3P,/>P,

1.30 1.20 0.99 1.29 1.28

1.96 1.61 1.32 1.80 1.39

Total/Total

1.17

1.59

blood or in the prevalence and density of the parasitaemia in the infants. Antibody levels in cord blood actually dropped in the wet season and the prevalence and density of parasitaemia in infants remained constant or rose in the period August to November. Table II shows the results of the 1977 wet season survey of pregnant and non-pregnant women by parity groups. Table III shows the results of the survey of pregnant women by week of gestation. Table IV shows the ratio of malaria prevalence and density in pregnant compared with non-pregnant matched for parity which was the best guide available to us for age. Table V shows the results of the immunofluorescence tests on pregnant and nonpregnant women in Mansakonko. It will be seen that pregnant women showed higher prevalence and density of parasitaemia throughout and that the ratio was fairly constant for the parity groupings. Among pregnant women prevalence and density of parasitaemia was highest in the Q 24 week of pregnancy group and declined thereafter. Dividing the women into those with IFT titres 1:320 and below and those with 1:640 and above, more pregnant women occurred in the lower titre group than non-pregnant women and among pregnant women the gestational group Q 24 weeks showed the greatest proportion below a titre of 1:320. Finally, Table VI shows prevalence and density of parasitaemia in three groups: 33 to 36 weeks of pregnancy, 37 to 40 weeks of pregnancy and nonpregnant women with infants less than one month old. The figures do not show any increase in prevalence or density in the puerperium. Discussion The results from pregnant women reported here show a seasonal distribution of malaria in the

o/oPrev. 50.4 39.1 35.8

of gestation

Density

log (n + 1)

2 x S.E.

1.22 0.85 0.73

0.17 0.12 0.14

women and consequent seasonal variation in their antimalarial immunoglobulin, as has also been shown for Sengalese children by REY et al. (1972). Antimalarial IgG in cord blood showed a seasonal variation but this did not follow the same pattern as in pregnant women. During the wet season, when parasitaemias and antibody levels rose sharply in pregnant women, antibody levels in cord blood dropped. Using fluorescein tagged anti-IgG and anti-IgM it could be shown that IgM did not contribute significantly to the antibody rise in pregnant women and its presence or absence could not explain the wide divergence in antimalarial immunoglobulin levels in pregnant women as compared to cord blood during the high transmission season. (May: Ig 2.05, IgG 1.93, IgM 0.28; August: Ig 3.02, IgG 2.81, IgM 0.35). It would seem possible that the malarial placenta acts as a barrier to the passage of antimalarial IgG from mother to foetus. CORR~A et al. (1978) in Senegal have noted positive IFT in mothers whose new-born offspring have a negative IFT. This inhibition of passage of antimalarial IgG may be having an effect, as the over-all prevalence and density of malaria parasites in all infants (nought to six months) and, specifically, the prevalence and density in the two to three-month age group remained steady or rose from August to November when prevalence and density of parasitaemia in pregnant mothers were dropping. The figures would support the hypothesis that the infant nourished from the infected intervillous spaces of the placenta may, three months later, be at a disadvantage to malaria parasites. The infection of the placenta must be of considerable duration and late in the pregnancy to achieve significant lowering of antimalarial IgG in the new-born child. The results of the major survey of pregnant and non-pregnant women confirm that pregnancy in an immune Gambian population is accompanied by greater prevalence and density of malaria parasites. This phenomenon occurs regardless of parity and age, though primiparous pregnant women may be proportionately more susceptible to malaria than multiparous women. In as much as prevalence and parasite density decline from the middle of the second trimester to term, we agree with PINGOUD (1969) that increased susceptibility is not constant throughout pregnancy. Like KORTMANN (1972) in Tanzania, we have no evidence that susceptibility to malaria is increased during the puerperium as compared to the last month of pregnancy. Finally, malaria prevalence and parasite density during pregnancy are generally inversely related to the IFT (IgG) titre. It seems that the pregnant state may inhibit the host’s ability to mount an IgG

V-Results

: 0 2

:

;

: 1 3

it

:

2

3

1: 6

12

:4

(P) and non-pa?t 6 i 9 21

f

(P) and non-parasitized

para$ized

parasitize:

Non-pregnant: P N-P

N-P

pgnant:

i37

10 16

83 1011 1”; 2:

205

;:

of immunofluorescence

Pregnant in weeks of gestation > 24w. 25-32 w. i i 33-40 w. 0 2

Total

Non-pregnant

Pregnant Primip I?1 fP2 > P3 Total

Table

3:

i:

zed (N-P)

2

(N-P)

13 29 11

t::

15 15

z;

2:

tests

18 42

13 33

10 24 12

:80

20 22

E

14 15

14 29

8 23

4;

14 19

9 8 14 31

6 19

3 11

2::

8 12

7 14

f

6 7

2 9

1;

53

3 3 5 11

:

:

i

2 1

:

3 1

0 0

:

:

00

: 1 2 = = = =

0.887 1.035 0.867 0,935

0.917 0.388 0.600 0.571

45139 = 1.154 100/l 16 = 0.862

28147 = 0.596 21/144= 0.563

21153 = 0.396 571105 = 0.543 31/33 = 0.939

47153 59157 39.45 145/155

33/36 = 31/80 = 45175 = 109/191=

2.70 2.60

2.47 2.50

2.37 2.51 2.57

2.62 2.64 2.63 2.63

2.61 2.39 2.52 2.49

0.57 0.58

0.64 0.61

0.62 0.61 0.63

0.57 0.60 0.58 0.58

0.70 0.56 0.61 0.62

0.06 0.04

0.07 0.04

0.07 0.05 0.08

0.06 0.06 0.06 0.03

0.08 0.05 0.06 0.44

R.S.BRAYAND

M.J.

431

ANDERSON

Table VI-Prevalence and density of P. falciparum malaria in pregnant women: the thiid trimester of pregnancy and the puerperium Category

Number

33-36 weeks pregnant 37-40 weeks pregnant Non-pregnant, child Q lm.

182 92 58

response to malaria parasitization and that this is most evident early in pregnancy. CORR~ et al. (1978) also noted a lower positive IFT prevalence in the period nought to 28 weeks as compared to the prevalence in the 28 to 35-week period. However, they also noted a lower prevalence in the period 35 to 42 weeks. As decreased levels of IgG specific for falciparum antigen are coincident with increased malaria prevalence and parasite density during pregnancy in general, and early pregnancy in particular, in our study a causal relationship might be inferred.

References Bruce-Chwatt, L. J. (1952). Malaria in African infants and children in Southern Nigeria. Annals of Tropical Medicine and Parasitology, 46, 173200. Clark, H. C. (1915). The diagnostic value of the placental blood film in aestivo-autumnal malaria. Journal of Experimental Medicine, 22, 427-444. Correa, P., Roffi, J., Diadhiou, F., Lauroy, J. & Bah, M. D. (1978). L’immunofluorescence indirecte dans l’association paludisme et grossesse chez la femme senegalaise. Bulletin de la Socie’te’ de Medicine Africaine Noire de la Langue Francaise 23, 141-148. Gamham, P. C. C. (1938). The placenta in malaria with special reference to reticula-endotheiial immunity. Transactions of the Royal Society of Tropical Medicine and Hygiene, 32, 13-34. Gilles, H. M., Lawson, J. B., Sibelas, M., Voller, A. & Allan, N. (1969). Malaria, anaemia and pregnancy. Annals of Tropical Medicine and Parasitology, 65, 245-263.

and recently

pregnant

POS.

% Prev.

Mean parasites/mms

;: 18

38.5 30.4 31.0

563 1514 324

Kortmann, H. F. C. M. (1972). Malaria and Pregnancy. Utrecht, The Netherlands : Drukkerij Elinkwiik. McGregor, I. A. &Wilson, R. J. M. (1971). Precipitating antibodies and immunoalobulins in P. falciphrum infections in the -Gambia, West Africa. Transactions of the Royal Society of Tropical Medicine and Hygiene, 65, 136-145. Made&, 0. & Kretschmar, W. (1966). Haiifigkeit und Ursachen der angeborenen Malaria in WestAfrica. Zeitschrsft fiir Tropenmedizin und Parasitologie, 17, 195210. Pingoud, E. (1969). Malariaplasmodien im Blute von Schwangeren tmd Nichtschwangeren in Abeokuta (West-Nigeria). Zeitschrift fiir Tropenmedizin und Parasitologic, 26, 279-287. Rey, M., Monjour, L., Gentilini, M. & Sow, A. (1972). Variations saisonnieres des anticorps fluorescents en zone d’hyperendemie palustre instable. Bulletin de la Sock%! de Pathologic Exotique, 65, 808-814. Seitz, H. M. (1971). Vergleich der Antikorpertiter gegen Plasmodium falciparum in miitterlichem Blut und in Nabelschnurblut. Zeitschrift fur Tropenmedizin. und Parasitologic, 22, 133-137. Williams, A. I. 0. & McFarlane, H. (1970). Immunoglobulin levels, malarial antibody titres and placental parasitaemia in Nigerian mothers and neonates. African Journal of Medical Science, 1,

369-376.

Accepted for publication

22nd January,

1979.