Malaria congenital infection constitutes a major risk in malaria endemic areas. In this study, we report the prevalence of transplacental malaria in Burkina Faso. In labour and delivery units, thick and thin blood films were made from maternal, placental, and umbilical cord blood to determine malaria infection. A total of 1,309 mother/baby pairs were recruited. Eighteen cord blood samples (1.4%) contained malaria parasites
In malaria high-transmission areas, some population groups are at considerably higher risk of infection with
Placental malaria is defined as the accumulation of
Vertical transmission of malaria from mother to foetus through the placenta and umbilical cord is defined as umbilical cord blood parasitemia. The transplacental transmission of
The direct burden of neonatal malaria infection in terms of prevalence is not well-described in malaria endemic areas. In fact, the method used to identify congenital transmission is peripheral blood of newborns or umbilical cord blood. Malaria parasites have been detected only rarely in the peripheral blood of newborns, whether the blood specimen is collected at the time of birth or hours later [
Studies published so far have documented contradictory levels of this burden. In countries without endemic malaria, congenital malaria has occurred in children born to women who have immigrated from malarious areas. However, transplacental transmission of
In Burkina Faso, the real prevalence of neonatal malaria is unknown but is estimated to be even higher. This assessment is based on presumptive malaria diagnosis. The present study was designed to determine the real burden of transplacental transmission, the risk factors associated with transplacental transmission, and the prevalence of cord blood and placental malaria parasitemia in malaria holoendemic area of Burkina Faso.
These results represent a pooled analysis of studies on malaria prevention in pregnant women [
The first study took place within six delivery units (DUs) of the Koupela health district, which is located around 120 km east of Ouagadougou. The second study was conducted in one DU of the Health District of Bousse. Both study sites have been extensively described elsewhere [
The study participants were pregnant women who voluntarily consented to participate in trials of malaria prophylaxis during pregnancy. They were encouraged to deliver at the health facility where the study samples were collected for processing.
Women delivering at the health facility, after giving informed consent, were asked a standard series of questions focused on sociodemographic characteristics, history of fever, antimalarial drug use, and the use of antimalarial chemoprophylaxis and bednets. Capillary blood was obtained by fingerstick for malaria blood film preparation. Placental blood films were prepared by identifying the maternal side of the placenta, wiping away excess blood, cutting into the surface, and placing pooled blood onto a slide. Umbilical cord blood samples were obtained by wiping away excess blood from a clamped cord, piercing it with a lancet, and placing a drop of expressed blood on a slide.
A detailed clinical examination was done on each newborn infant within 24 hours of delivery. Neonates were weighted with an electronic digital scale (±10 grams) (Tanita Corporation, Tokyo, Japan). The Dubowitz scoring system was used to estimate gestational age, using findings from physical and neurologic examinations. Scoring by APGAR index was performed at delivery but was not recorded in this study.
All blood films (maternal, placental, and cord) were stained with Giemsa and examined for parasites at the “Centre National de Recherche et de Formation sur le Paludisme” immunoparasitology laboratory in Ouagadougou. For thick films, parasites and leukocytes were counted in the same fields until 500 leukocytes were counted. Parasite densities were estimated using an assumed leukocyte count of 8,000 leukocytes/
Blood film results were considered to be positive if any asexual-stage parasites were identified and negative if no parasites were seen in 100 high-power fields.
Prematurity was defined as a gestational age <37 weeks as estimated by Dubowitz examination.
We defined low birth weight as a birth weight of <2,500 g with a gestational age ≥37 weeks.
Data collected in the study questionnaire were verified, then double-entered, and validated with EpiInfo, version 6.04 fr (Centers for Diseases Control and Prevention, Atlanta, USA). Data were analyzed using Epi-info and Stata 7.0. The analysis included data from births of all enrolled participants who delivered at the study centre clinic for whom data were available. Continuous, normally distributed data were described by mean and standard deviation and nonnormally distributed data by the median or geometrical mean and range. Proportions were compared using the Chi-square test and normally distributed continuous variables were compared using Student’s
The study was discussed with health authorities and local leaders to obtain their assent. The study was reviewed and approved by the Burkina Faso Ministry of Health ethical committee. Informed consent (by signature or thumbprint) was obtained after the consent document was read to the women in the local language. For illiterate mothers, the informed consent discussion process was witnessed by an impartial individual.
In total, 1,374 women delivered at the health unit; 1,309 women delivered singleton infants and provided data in the form of peripheral blood smear, placental blood, and umbilical cord blood. The profile of enrolled women is summarized in Table
Characteristics of enrolled women.
Gravidity | ||||
Characteristics | 1 | 2 | >2 | All |
Age, median (years) | ||||
Mean gestational age (weeks) | ||||
Mossi ethnic group (%) | 95.8 | 96.2 | 94.7 | 95.5 |
Sleeps under bednets (%) | 22.8 | 28.1 | 19.4 | 22.4 |
Mother death ( | 2 | 0 | 0 | 0 |
Pregnancy outcome | ||||
Stillbirth ( | 19 | 8 | 23 | 50 |
Miscarriage ( | 8 | 3 | 7 | 18 |
Weight (g) | ||||
Number of live babies ( | 460 | 270 | 579 | 1309 |
LBW* ( | 111 (24.1%) | 39 (14.4%) | 57 (9.8%) | 207 (15.8%) |
LBW*: low birth weight.
Primigravidae comprised 35% of the total sample size, and primigravidae and secundigravida together made up 56% of the total sample size. The mean duration of gestation was
Two women died during delivery as a consequence of eclampsia. Other complications during the delivery included peripartum haemorrhage, placental retention, and retroplacental haemorrhage. Fifty stillbirths (3.6%) and 18 (1.3%) miscarriages occurred at the DUs.
The weight range of babies was 700 g to 4,700 g with a mean weight of
Over one-third of the women were parasitaemic. Indeed, of the 1,309 pregnant women,
The proportion of newborns with
Gravidity | ||||
1 | 2 | >2 | All | |
Peripheral parasitemia (%) | 34.3 | 35.2 | 20.3 | 28.2 |
Geometric mean (parasites/ | 2,143.18 | 894.71 | 495.09 | 1,106.47 |
Negative peripheral parasitemia (%) | 32.4 | 18.8 | 48.8 | 71.7 |
Peripheral parasitemia 1–999 (%) | 30.9 | 29.5 | 39.6 | 11.4 |
Peripheral parasitemia 1,000–4,999 (%) | 53.3 | 19.6 | 27.2 | 7.1 |
Peripheral parasitemia 5,000–9,999 (%) | 52.5 | 22.5 | 25.0 | 3.1 |
Peripheral parasitemia ≥10,000 (%) | 58.8 | 25.0 | 16.2 | 5.2 |
Placental parasitemia (%) | 28.5 | 21.9 | 11.4 | 19.5 |
Geometric mean (parasites/ | 1384.55 | 396 | 591.71 | 830.52 |
Negative placental parasitemia (%) | 30.9 | 20.0 | 49.1 | 19.9 |
Placental parasitemia 1–999 (%) | 46.5 | 26.4 | 27.1 | 10.1 |
Placental parasitemia 1,000–4,999 (%) | 51.2 | 26.8 | 22.0 | 3.2 |
Placental parasitemia 5,000–9,999 (%) | 50.0 | 29.2 | 20.8 | 1.9 |
Placental parasitemia ≥10,000 (%) | 69.7 | 6.1 | 24.2 | 2.6 |
Umbilical cord parasitemia (%) | 2.4 | 1.5 | 0.5 | 1.4 |
Geometric mean (parasites/ | 263.74 | 644.08 | 222.16 | 315.69 |
Negative umbilical cord parasitemia (%) | 34.9 | 20.5 | 44.7 | 98.6 |
Umbilical cord parasitemia 1–999 (%) | 61.5 | 15.4 | 23.1 | 1.0 |
Umbilical cord parasitemia 1,000–4,999 (%) | 0.0 | 100 | 0.0 | 0.1 |
Umbilical cord parasitemia 5,000–9,999 (%) | 0.0 | 0.0 | 0.0 | 0.0 |
Umbilical cord parasitemia ≥10,000 (%) | 66.7 | 33.3 | 0.0 | 0.2 |
Umbilical cord blood parasitemia with
Out of 369 (28.2%) women with peripheral positive parasitemia, 211 (57.2%) had positive placental malaria and 14 (3.8%) had malaria parasites in the umbilical cord blood. Eleven (5.2%) of the 211 (57.2%) women with positive placental smears for malaria had malaria parasites in the umbilical cord blood. Of the 18 babies infected with
The number of pregnant women with peripheral, placental, and umbilical infection and high peripheral parasitemia (>10,000 parasites/
The combination of being born to a mother with both peripheral and placental infection conferred a greater likelihood of infection of the baby (61%). Being born with only maternal peripheral infection reduced the likelihood of having umbilical cord parasitemia (17%). There was an association between umbilical cord blood infection and maternal peripheral infection (
Univariate logistic regression of risk factors associated with
Proportion of newborns with umbilical cord blood parasitemia and the risk factor specified | Odds ratio | 95% confidence intervals | ||
---|---|---|---|---|
Peripheral positive parasitemia | 77.8% | 9.24 | 3.02–28.28 | <0.001 |
Peripheral parasitemia 1–999 | 22.2% | 2.23 | 0.75–7.20 | 0.13 |
Peripheral parasitemia 1,000–4,999 | 11.1% | 1.64 | 0.37–7.28 | 0.50 |
Peripheral parasitemia 5,000–9,999 | 22.2% | 9.62 | 3.01–30.69 | <0.001 |
Peripheral parasitemia ≥10,000 | 22.2% | 5.29 | 1.69–16.52 | 0.004 |
Placental positive parasitemia | 72.2% | 10.74 | 3.79–30.42 | <0.001 |
Placental parasitemia 1–999 | 16.7% | 1.78 | 0.50–6.24 | 0.36 |
Placental parasitemia 1,000–4,999 | 22.2% | 9.62 | 3.01–30.69 | <0.001 |
Placental parasitemia 5,000–9,999 | 5.6% | 3.13 | 0.40–24.56 | 0.27 |
Placental parasitemia ≥10,000 | 11.1% | 4.91 | 1.08–22.28 | 0.03 |
Primigravid | 61.1% | 2.96 | 1.13–7.69 | 0.02 |
Prematurity (<37 weeks) | 33.3% | 1.58 | 0.52–4.25 | 0.36 |
ITN | 33.3% | 1.69 | 0.63–4.56 | 0.29 |
LBW* | 27.8% | 2.10 | 0.74–5.97 | 0.16 |
Female sex of infant | 61.1% | 1.78 | 0.68–4.63 | 0.33 |
LBW*: low birth weight.
Being born to a mother with maternal peripheral parasite density ≥5,000 parasites/
We further examined the correlation between risk factors and the presence or absence of umbilical cord parasitemia to determine possible factors that might affect infection with malaria parasites in umbilical cord blood. Univariate logistic regression analysis indicated that being born with maternal parasitemia, being born with maternal parasite density ≥ 5,000 parasites/
Low maternal peripheral and placental parasitemia, prematurity, use of impregnated bednets, low birth weight, and female sex were independent risk factors for infection in the babies.
On multivariate logistic regression, none of the seven factors that were significant on univariate logistic regression remained significant (Table
Multivariate logistic regression of risk factors, associated with
Odds ratio | 95% confidence intervals | ||
---|---|---|---|
Peripheral positive parasitemia | 2.43 | 0.50–11.66 | 0.26 |
Peripheral parasitemia 5,000–9,999 | 3.71 | 0.92–14.98 | 0.06 |
Peripheral parasitemia ≥10,000 | 1.60 | 0.38–6.66 | 0.51 |
Placental positive parasitemia | 3.30 | 0.73–14.87 | 0.11 |
Placental parasitemia 1,000–4,999 | 2.27 | 0.58–8.77 | 0.23 |
Placental parasitemia ≥10,000 | 0.95 | 0.15–5.69 | 0.95 |
Primigravid | 1.91 | 0.70–5.21 | 0.20 |
This study shows the burden of malaria in pregnant women at the delivery unit in a malaria-endemic area of Burkina Faso. This investigation demonstrated the burden of malaria during pregnancy. The maternal peripheral parasitemia rate found in this study was 28.2%. This is similar to the rate reported in other countries in sub-Saharan Africa [
The prevalence of placental parasites was 19.5%. Previously reported prevalence rates of placental parasitemia in Africa have been highly variable, ranging from 17.2% to 57% [
The malaria rate in umbilical cord blood was low in the study population, occurring in 1.4% of all newborns, in 5.2% of babies born from mothers with peripheral and placental malaria infection, and in 5.1% of babies born from women with placental malaria infection only. Previously reported rates from different parts of Africa are highly variable, ranging from 0% to 54% [
Some studies have found high rates of transplacental malaria infection. The difference between those studies and our report could be explained by several factors. First, the high efficacy of malaria prophylaxis during this study may have allowed the women to clear their parasitaemia and therefore prevent placental and transplacental transmission of the infection to their babies. Second, our sole use of microscopy may have led to underdiagnosis, since this method has lower sensitivity than the newer molecular methods, such as real-time PCR, used in other studies [
Univariate analysis of the association of risk factors with the presence or absence of umbilical cord parasitemia revealed seven risk factors that was associated with
Maternal peripheral blood parasitemia was associated with umbilical cord blood parasitemia (
There were also associations between placental and cord blood parasitemia (
The density of maternal peripheral parasitemia (>5,000 parasites/
Finally, univariate analysis showed that being born from a first pregnancy was also linked to umbilical cord blood parasitemia (
Our data indicate that the rate of mother-to-child transmission of malaria, defined as positive umbilical cord blood parasitemia, appears to be low. Clinicians are therefore advised to investigate other aetiologies of fever in neonates. The low level of parasitaemia in cord blood suggests that contamination probably occurs during the labour period. Maternal, placental, and high density parasitemia were all associated with umbilical cord parasitemia. Prevention of malaria during pregnancy with effective antimalarial drugs should reduce the risk of infection for newborns.
The authors declare that they have no conflict of interests.
The authors express our gratitude to the pregnant women who participated in the study for their kind cooperation. Special thanks are due to all the staff of the health districts of Boussé and Koupela. The authors also thank the lab staff of the Centre National de Recherche et de Formation sur la Paludisme, Mr. Ouedraogo Z. Amidou for data management and Convelbo Nathalie for organizing the logistics of the study (Ouagadougou, Burkina Faso).