Epidemiology of malaria in pregnancy in central India.

Epidemiology of malaria in pregnancy in central India N. Singh,1 M.M. Shukla,2 & V.P. Sharma3 Analysis of three years of data from a malaria clinic operated by the Indian Council of Medical Research (ICMR) in the Government Medical College Hospital in Jabalpur, central India, showed a high malaria prevalence among pregnant women, which was statistically highly significant (P <0.0001) compared with the situation among nonpregnant women. Cerebral malaria was a common complication of severe Plasmodium falciparum infection, with a high mortality during pregnancy, requiring immediate attention. The study also showed that malaria infection was more frequent in primigravidae, falling progressively with increasing parity. Mean parasite densities were significantly higher in pregnant women compared with nonpregnant women for both P. falciparum (P <0.001; df =137) and P. vivax (P <0.05; df = 72) infection. Pregnant women with falciparum or vivax malaria were significantly more anaemic than noninfected pregnant women or infected nonpregnant women. The average weight of 155 neonates from infected mothers was 350 g less than that of 175 neonates from noninfected mothers. This difference in birth weight was statistically significant for both P. falciparum (P <0.0001; df = 278) and P. vivax (P <0.0001; df = 223) infection. Congenital malaria was not recorded. We conclude that pregnant women from this geographical area require systematic intervention owing to their high susceptibility to malaria during pregnancy and the puerperium. Voir page 571 le re´sume´ en franc¸ais. En la pa´gina 572 figura un resumen en espan˜ol. Introduction It is generally agreed that the prevalence of malaria is higher among pregnant women than other groups (1), and that this can lead to abortion, intrauterine fetal death, premature delivery and even maternal death (2). The greater part of our knowledge about malaria in pregnancy is derived from studies carried out in tropical Africa, which show differences in the clinical epidemiological pattern of malaria in preg- nancy from one endemic setting to another (3). Systematic studies from malaria-endemic regions in India are lacking except for those carried out in Chandigarh (4, 5) and Surat, Gujarat (6), two areas that differ in climate, intensity of malaria transmis- sion, use of malaria control measures, and socio- cultural attitudes towards the disease. In the absence of such a study in central India, we examined the relationship between malaria infection and pregnancy in Jabalpur — information which would help to develop control strategies. This endemic region is of special interest because the population is exposed to malaria from both Plasmodium vivax and P. falciparum. Materials and methods Study area Jabalpur district (area, 10 160 km2) in central India (Madhya Pradesh), has a mixed rural, tribal, and urban population (total, ca. 2.2 million). The district is mostly rocky with an undulating terrain and has no proper drainage and irrigation facilities. The study was carried out in the Government Medical College, which is surrounded by typical urban slums; the river Narmada, about 5 km away, supports the breeding of three primary vector species — Anopheles culicifacies, A. fluviatilis and A. stephensi. The Medical College is the largest medical facility in the district and serves as both a hospital for local people and a referral hospital for the adjoining six districts. Both P. vivax and P. falciparum malaria are common in the area (7), and there is a definite seasonal trend (8). Cases are mainly due to P. vivax in the dry hot season (March–June) and to P. falciparum during the monsoon and post- monsoon period. Resistance of P. falciparum to chloroquine is common (9, 10). Antenatal clinic The obstetrics and gynaecology department of the Medical College has a busy antenatal clinic, which an average of 500 women attend every month. Pregnant women usually first attend the antenatal clinic in the fourth or fifth month of gestation and make three or four visits before delivery. At enrolment, a clinical and obstetric history is recorded for each woman by the medical officer, followed by a full examination, including measuring the woman’s weight, tempera- ture, pulse, blood pressure, fundal height, fetal heart 1 Deputy Director, Malaria Research Centre (Field Station), Medical College Building, Jabalpur 482 003, India. Requests for reprints should be sent to this author. 2 Senior Research Officer, Malaria Research Centre (Field Station), Jabalpur, India. 3 Director, Malaria Research Centre, 20 Madhuvan, Vikas Marg, Delhi, India. Reprint No. 0010 567Bulletin of the World Health Organization, 1999, 77 (7) # World Health Organization 1999 rate, and determining the presence of oedema or anaemia. An estimate of gestational age is also recorded, based on the fundal height and fetal size. The majority of women attend the antenatal clinic for a routine check-up, but those with fever or a history of fever at some time during the pregnancy are referred to the malaria clinic in the Malaria Research Centre (Indian Council of Medical Research). Fever cases among those attending other departments in the hospital with minor complaints, or while visiting patients or accompanying patients to the hospital, are also referred to the malaria clinic. There were over 4000 deliveries in 1992 at the Medical College and the number has been increasing steadily. Women attending the antenatal clinic are given one month’s supply of iron(II) sulfate tablets (60 mg daily) and folic acid tablets (5 mg daily). Deliveries are usually conducted by nurses and the birth weights are recorded within 24 hours by one of the clinic staff. Over 85% of the obstetric admissions to the hospital lived within a few miles of the hospital; about 5% were women from surrounding villages who arrived only when labour was imminent, and about 10% were referred from nearby districts. Study population The study was carried out in 1992–95 on women with fever or a history of fever, including pregnant women from 12 weeks’ gestation up to 40 days after delivery. All the women belonged to the lower socioeconomic groups and worked in their homes and fields until term. Personal and reproductive histories were obtained from all the study subjects, and they were asked about any treatment taken for malaria during the pregnancy. Their answers were not always clear, many women being unaware of the names or nature of the drugs they had received. Thick and thin films of fingerprick blood collected from the women were stained with Giemsa and examined under the microscope. On first presentation each case was classified as new and on subsequent presentations as old, and their parasitic findings were recorded separately. Only the first parasitic (febrile) episode of malaria was considered for data analysis and parasite counting. All 40 cases referred from neighbouring districts were excluded from the study because their personal and reproductive records were not complete; they often reported at a late stage in a critical condition, and the majority of these women died within 96 hours of admission to hospital (11). There were two control groups in the study. The first group, chosen for assessment of the prevalence of vivax / falciparum malaria and parasite density, consisted of nonpregnant women of repro- ductive age suffering from malaria who were attending the hospital or malaria clinic for minor complaints during the same period. The second group, chosen to compare anaemia and low birth weight, were pregnant women from the study group who had fever but no malaria infection. These controls were matched with infected pregnant women for parity, since few women knew their ages. Thick blood smears were examined for parasite counts (per mm3) using standard methods (12). The women were treated with chloroquine (1500 mg in divided doses), as recommended by the national malaria eradication programme. The results of routine laboratory investigations (complete haemogram, blood groups, etc.) were taken from the hospital records. Haemoglobin was determined using the cynanomethaemoglobin method; a haemoglobin level <10 g/dl was con- sidered low. The birth weights of neonates were also obtained from the hospital records; low birth weight Table 1. Malaria parasitaemia, anaemia during pregnancy, and low birth weight among the study subjects in central India All Cases with Cases with Controls 1a Controls 2a casesa P. vivax P. falciparum infection infection No. tested 2127 365 365 1 984 No. selected 365 (17%) 121 (33%) 244 (67%) 150 (8%) 1 762 (i.e. 2127–365) Pregnant Yes Yes Yes No Yes Fever Yes Yes Yes Yes Yes Malaria Yes Yes Yes Yes No No. with Hb 271 (74%) 83 (69%) 188 (77%) 85 (57%) 88 (5%) data available Mean Hb + SD — 9.05 + 1.39 6.42 + 1.98 9.68 + 1.43 10.03 + 1.11 (g/dl) No. with birth 155 (42%) 50 (41%) 105 (43%) — 175 (10%) weight available Mean wt + SD 2.18 + 0.25 2.22 + 0.30 2.15 + 0.21 — 2.53 + 0.43 (kg) a Cases are pregnant women with fever. Controls 1 = infected nonpregnant women. Controls 2 = noninfected pregnant women. Research 568 Bulletin of the World Health Organization, 1999, 77 (7) was defined as <2.5 kg. The presence of placental infection and cord blood analyses could not be included in the present study because these investiga- tions were not carried out in the hospital. Blood smears were prepared from all neonates (with and without fever) whose mothers were enrolled in the study. All neonates for the study of low birth weight were full-term deliveries (twins excluded). The results were analysed according to the type of malaria infection, parasite density in peripheral blood, parity, and period of gestation. Differences in means were tested by Student’s t- test and differences in proportions by the Z-test. The study was approved by the Ethical Committee of the Malaria Research Centre. Results Table 1 shows that of 2127 pregnant women, 121 were infected with P. vivax (33%) and 244 with P. falciparum (67%), of whom 17 were cerebral malaria cases. Among 1984 nonpregnant women with fever, there were only 115 falciparum and 35 vivax malaria cases. The slide positivity rate (SPR) and slide falciparum rate (SFR) were significantly higher (P <0.001) in pregnant than nonpregnant women. Table 2 shows the distribution of malaria cases by trimester and parity; 121 (23%) of 527 women in their first pregnancy (P0) were parasitaemic, com- pared with 184/1075 (17%) in their second and third pregnancies (P1-2) and 60/525 (11%) in their fourth (or more) pregnancy. Overall, the mean prevalence of vivax malaria was highest during the second trimester (10%) and lowest during the puerperium (2.5%). However, the mean prevalence of falciparum malaria was nearly the same in all trimesters and the puerperium. Parasite density was counted in only 213 pa- tients out of 515 (pregnant and nonpregnant) because 220 of the thick blood smears were not prepared correctly, 70 were poorly stained, and 12 slides were missing. Women of all parities were therefore grouped together in subsequent analyses of pregnant and nonpregnant women. Mean parasite densities were significantly higher in all three trimesters and the puerperium for the pregnant women, compared with nonpregnant women, for both P. falciparum (P <0.0001) and P. vivax (P <0.05) malaria. However, the parasite densities in the three trimesters (all parities combined) did not show any statistically significant differences between trimesters for both types of infections. The women with falciparum malaria were significantly more anaemic than the noninfected pregnant women (P <0.0001; df = 274) or the infected nonpregnant women (P <0.001; df = 271); the women with vivax malaria were also significantly more anaemic than the noninfected pregnant women (P <0.0001; df = 169) or infected nonpregnant women (P <0.005; df =166). Furthermore, among pregnant women (all parities combined), those infected with P. falciparum were significantly more anaemic than those with P. vivax (P <0.001; df = 269). However, among nonpregnant women no such difference between P. falciparum and P. vivax was observed. When the association between malaria infection and anaemia was stratified by parity, primigravidae with P. falciparum were more anaemic than second gravidae and multigravidae, although only with the latter was the difference statistically significant (P <0.005; df = 97). However, in vivax- infected pregnant women or noninfected pregnant women, the anaemia increased with increasing gravidae, but this was not statistically significant. Three maternal deaths (two primigravidae and one multigravida) occurred in women who were infected with P. falciparum during the late third trimester (8–9 months). Two cases with P. falciparum Table 2. Malaria parasitaemia by parity and stage of pregnancy Stage of pregnancy Puerperium Total 1st trimester 2nd trimester 3rd trimester Cases Pva Pfb Cases Pva Pfb Cases Pva Pfb Cases Pva Pfb Cases Pva Pfb Parity P0 39 6 7 132 16 20 317 16 42 39 1 13 527 39 82 P1-2 50 3 5 168 20 17 505 32 59 352 9 39 1075 64 120 P3+ 35 2 4 77 2 10 275 11 15 138 3 13 525 18 42 Total 124 11 16 377 38 47 1097 59 116 529 13 65 2127 121 244 Mean parasite density + SDc Pva 4081.4 + 4290.5 14 230 + 19 050 14 401 + 17 176 8233 + 11 103 1780 + 1447 (6)d (26) (21) (7) (14) Pfb 11 435 + 17 269 7805 + 8087 10 882 + 14 607 16 661 + 22 499 3931 + 7419 (12) (25) (47) (19) (36) a Cases with P. vivax infection. b Cases with P. falciparum infection. c All parities combined. Data provided by 213 ‘‘good thick smears’’ (pregnant 137; puerperium 26; nonpregnant 50). d Figures in parentheses indicate the number of cases providing parasite density. Epidemiology of malaria in pregnancy in India 569Bulletin of the World Health Organization, 1999, 77 (7) infection died in the puerperium with anaemia (haemoglobin, 7.1 g/dl). Three abortions were recorded in women who were infected with P. falciparum during the first trimester (all primigra- vidae). Two stillbirths were recorded in women with P. falciparum (one primi- and one multigravida). Only one abortion and one stillbirth were recorded in vivax-infected primigravidae. Among noninfected pregnant women, only one abortion (primigravida) and one stillbirth (multigravida) were recorded. The clinical spectrum of 17 cerebral malaria cases (P0, 6; P1-2, 7; P3+, 4) was much more severe. All of them presented with a history of fever which was associated with neurological signs and haemorrhage (5.8%), jaundice with severe anaemia (11.7%), convulsions (29.4%), and unconsciousness (53%). Of these 17 women, 12 died (P0, 5; P1-2, 4; P3+, 3); among the 5 who recovered, one aborted and two had stillbirths. The mean weights of the newborn babies from the noninfected pregnant women and infected pregnant women with different types of infections (all parities combined) are shown in Table 1. The mean birth weight of the babies was significantly lower in the infected group than in the noninfected group (P <0.0001; df = 328); the mean birth weight of babies in the falciparum-infected group was lower than that in the vivax-infected group (P40.05; df = 153). With regard to low birth weight, 89% of babies weighed <2.5 kg in the infected group compared with 38% in the control group, which is highly significant statistically (P <0.0001; Z = 9.41). Peripheral blood smears taken from the new- born babies were negative for malaria parasites in all the cases. Discussion Since very little information is available on the relationship between malaria infection and pregnancy in the Indian subcontinent, the present hospital- based investigation on 365 pregnant women in central India is of special interest. The highest malaria prevalence was among primigravidae, followed by second gravidae and multigravidae, which accords with reports from Africa (3, 13). While almost nothing was known about the incidence of P. vivax malaria, the prevalence of P. falciparum malaria among pregnant women was highest early in the second trimester with a decline towards term, and the incidence postpartum was usually similar to the pre-pregnancy level reported for Africa (3, 12, 13). An exactly identical pattern was shown by P. vivax malaria in our study, while the P. falciparum prevalence was nearly the same in all trimesters. The high prevalence of P. vivax in the second trimester was probably due to relapses, while similar prevalences of P. falciparum in all trimesters indicate primary infections, which suggests that the pattern of malaria may vary in areas of different endemicity (3). A prospective controlled study in Chandigarh (5) showed that the parasite density was highest in the third trimester for both vivax and falciparum malaria, that parasite density remained high during the puerperium for both types of infections compared with the situation for non- pregnant women, and that parity had no influence on the parasite density. From our limited results, it is not possible to assess accurately the relationship between malaria parasite density and parity. Moreover, no conclusion can be drawn on the peak prevalence of parasitaemia for both parasite species since a much larger sample is needed to evaluate the effect of parity on parasite density. Anaemia is a major problem among pregnant women in this region. The mean haemoglobin level was lower in primigravidae infected with P. falciparum, as reported in other malarious areas (14–16). In contrast, anaemia usually increases in vivax-infected women and in noninfected pregnant women with parity, with primigravidae being the least anaemic group. This is presumably because iron deficiency tends to increase as gravidity increases (17). It is unfortunate that the haemoglobin data from our pregnant and nonpregnant study subjects were incomplete because many women feared the dis- comfort of blood sampling; we therefore do not have a more precise picture of the effect of anaemia in the infected and noninfected groups. Furthermore, since the etiology of anaemia in pregnancy is multifactorial, i.e. due to iron deficiency, folate deficiency and hookworm infection (18), the effect of malaria on this condition is difficult to ascertain. Quantification of the relative contributions of malaria, hookworm and malnutrition to anaemia is thus an important step in selecting appropriate control measures to prevent the adverse perinatal outcomes associated with anaemia in pregnancy (19). The association of low birth weight with malaria is well known (1), the mean birth weight of neonates born to infected mothers in our study being 350 g less than that of their counterparts born to noninfected mothers. This finding must be inter- preted with caution, however, because it is malaria infection of the placenta that is associated with low- birth-weight infants. Peripheral parasitaemia does not mean that there is a placental parasite infection (3); similarly, we cannot be sure that clearance of parasites from peripheral blood indicates the elim- ination of placental infection. Our findings also show that cerebral malaria is one of the common complications of severe falciparum malaria, with a high mortality during pregnancy. Furthermore, five maternal deaths, three abortions, and two stillbirths occurred among our uncomplicated P. falciparum cases. Unfortunately, all our referral cases had to be excluded because of incomplete records. Their omission from our analysis is likely to have underestimated the incidence of cerebral malaria and its complications during preg- nancy in the local population. Even though WHO (20) has recommended the use of an effective antimalarial throughout pregnancy Research 570 Bulletin of the World Health Organization, 1999, 77 (7) in an effort to prevent the adverse effects of malaria on the mother, fetus and newborn child, chemo- prophylaxis is not administered routinely by the national malaria eradication programme (NMEP) in India. The only recommendation by the Ministry of Health is treatment of febrile episodes. Thus, all fever cases were initially given chloroquine in divided doses as first-line treatment. Patients who could not tolerate oral administration were given chloroquine by intramuscular injection. Chloroquine-resistant cases received a single dose of sulfadoxine–pyr- imethamine. Cases resistant to all the above drugs, and cases of severe and complicated falciparum infection, were given intravenous quinine, as in- dicated in the NMEP schedule. The available data from this hospital-based study show a high prevalence of cerebral malaria and anaemia in pregnant women, indicating the need for malaria chemoprophylaxis. It should be routine policy to provide such chemotherapy to pregnant women through malaria clinics in endemic areas. This policy should be adopted despite the increased burden on limited medical budgets and adminis- trative difficulties, because protection of this group of women at risk deserves serious consideration. However, most pregnant women do not usually go to the hospital or clinic until the second trimester. Educational campaigns informing women of child- bearing age about the dangers of malaria in pregnancy and the potential benefits available to expectant mothers would help improve the rate of early attendance at the clinic. The study results also show the need for changes in the Medical College Hospital and malaria clinics, e.g. introduction of training/ workshops for all paramedical staff and laboratory technicians in order to obtain more reliable and precise data. However, the extent to which chemo- prophylaxis can reduce the risk of malaria during pregnancy will vary depending on the relative prevalences of vivax and falciparum malaria and on the level of drug resistance against P. falciparum. n Acknowledgements We are indebted to all members of the Medical College Hospital who assisted in this investigation. Special thanks are due to Dr J.P. Kapoor, Dean of the Medical College, and Dr R. Shrivastava, Professor and Head, Department of Obstetrics and Gynaecol- ogy, for allowing us to prepare blood smears from pregnant women who gave birth in the Medical College Hospital and from their neonates. We are grateful to Mr C.A. Thomas, Administrative Officer, Regional Medical Research Centre (ICMR), for help in typing the manuscript. Re´sume´ Epide´miologie du paludisme en cours de grossesse en Inde centrale L’analyse de trois ans de donne´es provenant d’un dispensaire antipaludique fonctionnant au centre hospi- talier universitaire de Jabalpur sous l’e´gide de l’Indian Council of Medical Research (Inde centrale), a mis en e´vidence une forte incidence de paludisme chez les femmes enceintes, incidence qui s’est re´ve´le´e tre`s significative du point du vue statistique (p <0,0001) comparativement a` la pre´valence enregistre´e chez les autres femmes. Les donne´es ont e´te´ analyse´es en fonction de la nature de l’infection palustre (a` falciparum ou a` vivax), du stade de la gestation et de la parite´ ainsi que de la densite´ parasitaire dans le sang pe´riphe´rique. On a confirme´ la pre´sence d’infections a` Plasmodium falciparum et a` P. vivax). L’acce`s pernicieux s’est re´ve´le´ eˆtre une complication fre´quente des infections graves a` falciparum, ne´cessitant une intervention imme´diate en raison de la forte mortalite´ qu’il entraıˆne chez les femmes enceintes. L’e´tude montre que les infections palustres sont plus fre´quentes chez les primipares et qu’elles se font de plus en plus rares a` mesure que la parite´ augmente. En ge´ne´ral, c’est pendant le second trimestre que la pre´valence du paludisme a` vivax a e´te´ la plus forte (10%) pour tomber a` son plus bas niveau pendant les suites de couches (2,5%). Dans le cas des infections a` falciparum en revanche, la pre´valence a e´te´ presque identique quel que soit le trimestre et pendant les suites de couches. La densite´ parasitaire moyenne e´tait sensiblement plus e´leve´e chez les femmes enceintes que chez les autres femmes, qu’il s’agisse de P. falciparum (p < 0,001; ddl = 137) ou de P. vivax (p < 0,05; ddl = 72). Les femmes enceintes porteuses d’une infection a` falciparum ou a` vivax e´taient sensiblement plus ane´mie´es que les femmes enceintes non impalude´es ou que les femmes non enceintes impalude´es. Par ailleurs, chez les femmes enceintes (toutes parite´s confondues), les femmes ayant une infection a` falciparum sont significativement plus ane´mie´es que celles ayant un paludisme a` P. vivax. Le poids moyen de 155 nouveau-ne´s de me`res impalude´es e´tait de 350 g infe´rieur a` celui des 175 nouveau-ne´s de me`res non impalude´es. Cette diffe´rence dans le poids de naissance e´tait statistiquement significative tant pour les infections a` P. falciparum (p < 0,0001; ddl = 278) que pour les infections a` P. vivax (p < 0,0001; ddl = 223). Les cas de paludisme conge´- nital n’ont pas e´te´ enregistre´s. Nous concluons de cette e´tude que dans la re´gion, les femmes enceintes doivent eˆtre syste´matiquement prises en charge du fait du risque e´leve´ de paludisme qu’elles courent pendant la grossesse et les suites de couches. Epidemiology of malaria in pregnancy in India 571Bulletin of the World Health Organization, 1999, 77 (7) Resumen Epidemiologı´a del paludismo entre las embarazadas en la India central En el Hospital Universitario Estatal de Jabalpur, en la India central, se analizaron datos reunidos durante tres an˜os en un consultorio de paludismo pertenenciente al Consejo de Investigaciones Me´dicas de la India. El estudio revelo´ una alta prevalencia de paludismo entre las mujeres embarazadas, estadı´sticamente muy signi- ficativa (p < 0,0001) en comparacio´n con la prevalencia entre las mujeres no embarazadas. Se analizaron los resultados segu´n el tipo de infeccio´n palu´dica (falcipa- rum o vivax), el periodo de gestacio´n y la paridad, y la densidad de para´sitos en sangre perife´rica. Se confirmo´ la prevalencia tanto de Plasmodium vivax como de P. falciparum. El paludismo cerebral era una complica- cio´n comu´n de la infeccio´n grave por falciparum, con una mortalidad alta durante el embarazo, y requerı´a atencio´n inmediata. El estudio mostro´ que la infeccio´n palu´dica era ma´s frecuente entre las primigra´vidas, disminuyendo de forma progresiva al aumentar la paridad. Globalmente, la prevalencia media de paludismo por vivax alcanzo´ valores ma´ximos durante el segundo trimestre (10%) y mı´nimos durante el puerperio (2,5%). Sin embargo, la prevalencia media de paludismo por falciparum fue casi ide´ntica en los distintos trimestres y en el puerperio. Las densidades medias de para´sitos fueron significativa- mente mayores en las mujeres embarazadas que en las no embarazadas, tanto en el caso de P. falciparum (p < 0,001; gdl = 137) como en el de P. vivax (p < 0,05; gdl = 72). Las embarazadas con paludismo por falciparum o vivax habı´an alcanzado un grado de anemia significa- tivamente mayor que el de las mujeres embarazadas no infectadas o las mujeres no embarazadas infectadas. Adema´s, entre las mujeres embarazadas (todas las paridades combinadas), las mujeres infectadas por falciparum habı´an llegado a estar significativamente ma´s ane´micas que las afectadas por vivax. El peso promedio de 155 recie´n nacidos de madres infectadas fue inferior en 350 g al de 175 recie´n nacidos de madres no infectadas. Esa diferencia en el peso al nacer fue estadı´sticamente significativa para ambos P. falciparum (p < 0,0001; gdl = 278) y P. vivax (p < 0,0001; gdl = 223). No habı´a informacio´n regis- trada sobre casos de paludismo conge´nito. Nuestra conclusio´n es que es necesario adoptar medidas sistema´ticas con las mujeres embarazadas de la zona estudiada, dada su alta susceptibilidad al paludismo durante el embarazo y el puerperio. References 1. Bruce-Chwatt LJ. Malaria in African infants and children in southern Nigeria. Annals of tropical medicine and parasitology, 1952, 46: 173–200. 2. Mvondo JL, James MA, Cambell CC. Malaria and pregnancy in Cameroonian women. Effect of pregnancy on Plasmodium falciparum parasitaemia and the response to chloroquine. Tropical medicine parasitology, 1992, 43: 1–5. 3. McGregor IA. Epidemiology, malaria and pregnancy. American journal of tropical medicine and hygiene, 1984, 33: 517–525. 4. Sholapurker SL, Gupta AN, Mahajan RC. Clinical course of malaria in pregnancy: a prospective controlled study from India. Transactions of the Royal Society of Tropical Medicine and Hygiene, 1988, 82: 376–379. 5. Sholapurker SL et al. Malarial parasite density in infected pregnant women from northern India. Indian journal of medical research, 1988, 88: 228–230. 6. Nair LS, Nair AS. Effects of malaria infection on pregnancy. Indian journal of malariology, 1993, 30: 207–214. 7. Singh N. Epidemiology of falciparum and vivax malaria in pregnancy in India. Paper presented at a Symposium on Malaria in Pregnancy in Kisumu, Kenya, organized by the Centers for Disease Control and Prevention (CDC), Atlanta, and the Kenyan Medical Research Institute (KEMRI), Kenya, 11–13 November 1997. 8. Singh N et al. Prevalence of malaria among pregnant and non-pregnant women of district Jabalpur, Madhya Pradesh. Indian journal of malariology, 1995, 32: 6–13. 9. Singh N et al. A focus of high-degree chloroquine-resistant P. falciparum in Mandla district (M.P.). Indian journal of malariology, 1989, 26: 45–51. 10. Singh N, Tyagi AK, Sharma VP. Drug-resistant Plasmodium falciparum in Mandla District, Madhya Pradesh. Indian journal of malariology, 1995, 32: 174–177. 11. Singh N, Shukla MM, Valecha N. Malaria parasite density in pregnant women of district Jabalpur, Madhya Pradesh. Indian journal of malariology, 1996, 33: 41–47. 12. Steketee RW et al. In-vivo response of Plasmodium falciparum to chloroquine in pregnant and non-pregnant women in Siaya district, Kenya. Bulletin of the World Health Organization, 1987, 65: 885–890. 13. Brabin BJ. An analysis of malaria in pregnancy in Africa. Bulletin of the World Health Organization, 1983, 61: 1005–1016. 14. Brabin BJ et al. Consequences of maternal anaemia on outcome of pregnancy in a malaria-endemic area in Papua New Guinea. Annals of tropical medicine and parasitology, 1990, 84: 11–84. 15. Nosten F et a.l Malaria in pregnancy in an area of unstable endemicity. Transactions of the Royal Society of Tropical Medicine and Hygiene, 1991, 85: 424–429. 16. Shulman CE et al. Malaria is an important cause of anaemia in primigravidae: evidence from a district hospital in coastal Kenya. Transactions of the Royal Society of Tropical Medicine and Hygiene, 1996, 90: 535–539. 17. Isah HS et al. Anaemia and iron status of pregnant and non-pregnant women in the guinea savanna of Nigeria. Annals of tropical medicine and parasitology, 1985, 79: 485–493. 18. The clinical management of acute malaria. New Delhi, World Health Organization Regional Office for South-East Asia, 1986 (WHO Regional Publications, SEARO Series, No. 9). 19. Brabin BJ. An assessment of low-birth-weight risk in primiparae as an indicator of malaria control in pregnancy. International journal of epidemiology, 1991, 20: 276–283. 20. WHO Expert Committee on Malaria. Eighteenth report. Geneva, World Health Organization, 1986 (WHO Technical Report Series, No. 735). Research 572 Bulletin of the World Health Organization, 1999, 77 (7)