Background: Despite that over 90 million pregnancies are at risk of Plasmodium vivax infection annually, little is known about the epidemiology and impact of the infection in pregnancy. Methodology and principal findings: We undertook a health facility-based prospective observational study in pregnant women from Guatemala (GT), Colombia (CO), Brazil (BR), India (IN) and Papua New Guinea PNG). Malaria and anemia were determined during pregnancy and fetal outcomes assessed at delivery. A total of 9388 women were enrolled at antennal care (ANC), of whom 53% (4957) were followed until delivery. Prevalence of P. vivax monoinfection in maternal blood at delivery was 0.4% (20/4461) by microscopy [GT 0.1%, CO 0.5%, BR 0.1%, IN 0.2%, PNG 1.2%] and 7% (104/1488) by PCR. P. falciparum monoinfection was found in 0.5% (22/4463) of women by microscopy [GT 0%, CO 0.5%, BR 0%, IN 0%, PNG 2%]. P. vivax infection was observed in 0.4% (14/3725) of placentas examined by microscopy and in 3.7% (19/508) by PCR. P. vivax in newborn blood was detected in 0.02% (1/4302) of samples examined by microscopy [in cord blood; 0.05% (2/4040) by microscopy, and 2.6% (13/497) by PCR]. Clinical P. vivax infection was associated with increased risk of maternal anemia (Odds Ratio-OR, 5.48, [95% CI 1.83–16.41]; p = 0.009), while submicroscopic vivax infection was not associated with increased risk of moderate-severe anemia (Hb<8g/dL) (OR, 1.16, [95% CI 0.52–2.59]; p = 0.717), or low birth weight (<2500g) (OR, 0.52, [95% CI, 0.23–1.16]; p = 0.110). Conclusions: In this multicenter study, the prevalence of P. vivax infection in pregnancy by microscopy was overall low across all endemic study sites; however, molecular methods revealed a significant number of submicroscopic infections. Clinical vivax infection in pregnancy was associated with maternal anemia, which may be deleterious for infant’s health. These results may help to guide maternal health programs in settings where vivax malaria is endemic; they also highlight the need of addressing a vulnerable population such as pregnant women while embracing malaria elimination in endemic countries.
The study protocol was reviewed and approved by the national and/or local ethics review boards in each of the study sites, the Institutional Review Board (IRB) of the Centers for Disease Control and Prevention (CDC), and the Hospital Clinic of Barcelona Ethics Review Committee. The study was conducted in accordance with the Good Clinical Practice Guidelines, the Declaration of Helsinki, and local rules and regulations of each partner country. This was a health facility-based prospective observational study of pregnant women attending routine antenatal care (ANC) clinics undertaken between June 2008 and October 2011 in five P. vivax endemic countries—Colombia (CO), Guatemala (GT), Brazil (BR), India (IN) and Papua New Guinea (PNG)–of different malaria endemicity characteristics (Table 1). The historic levels of malaria transmission varied across study sites from hypoendemic in Guatemala, India and Colombia, mesoendemic in Brazil, to hyperendemic in PNG. P. vivax was the predominant species in all sites except for PNG where P. vivax co-existed with P. falciparum (50%) and other species (P. malariae and P. ovale, 5%). Abbreviations: CQ: Chloroquine, PQ: Primaquine, SP: Sulfadoxine-pyrimethamine, QNN: Quinine, CLIN: Clindomycin, AL: Artemeter-Lumefantrine, ITNs: Insecticide-treated bednets, ADI: Active Detection of Infection, ACTs: Artemisin-combined therapy, PCD: Passive detection of cases. a No P. falciparum cases reported in the area since 2009. b Data from 2009. c 79% of women completed at least 1 ANC visit. d 150mg CQ base (tab 250 mg), 4 tab 1st day and 3 tab 2nd and 3rd day. e 500mg sulfadoxine/25mg pyrimethamine (3tab, monodosis). f QNN: 10mg/kg every 8h, CLIN: 10mg/kg every 12h. g AL: Artemeter (20mg)-Lumefantrine (120mg), twice per day over 3 days. h New protocol adopted in 2009. Uncomplicated P.falciparum malaria: AL. Uncomplicated P.vivax malaria: AL+PQ. i Indoor residual spraying and larvicides in general population. Source of data: Colombia: Montoya-Lerma J, Solarte YA, Giraldo-Calderón GI, Quiñones ML, Ruiz-López F, Wilkerson RC, González. Malaria vector species in Colombia: a review. Mem Inst Oswaldo Cruz 2011, 106:223–238. Chaparro P, Padilla J, Vallejo AF, Herrera S: Characterization of a malaria outbreak in Colombia in 2010. Malar J 2013, 12:330. Guía de atención médica para el diagnóstico y tratamiento de la malaria. Ministerio de Salud de Colombia. 2010. Guatemala: Padilla N, P. Molina, J. Juarez, D. Brown and C.Cordon-Rosales. Potential malaria vectors in northern Guatemala (Vectores potenciales de malaria in la region norte de Guatemala). J Am MosqControl Assoc 1992;8:307–8. Brazil: http://portalweb04.saude.gov.br/sivep_malaria/default.asp. Accessed 15 January 2017. India: Kochar DK, Sirohi P, Kochar SK, Budania MP, Lakhotia JP. Dynamics of malaria in Bikaner, Rajasthan, India (1975–2006). J Vector Borne Dis. 2007 Dec;44(4):281–4. PNG: Papua New Guinea National Department of Health: National Malaria Treatment Protocol. Port Moresby: National Department of Health ed., 1st edition; 2009. Marfurt J, Müeller I, Sie A, Maku P, Goroti M, Reeder JC, Beck HP, Genton B. Low efficacy of amodiaquine or chloroquine plus sulfadoxine-pyrimethamine against Plasmodium falciparum and P. vivax malaria in Papua New Guinea. Am J Trop Med Hyg. 2007 Nov;77(5):947–54. Barnadas C, Koepfli C, Karunajeewa HA, Siba PM, Davis TM, Mueller I. Characterization of treatment failure in efficacy trials of drugs against Plasmodium vivax by genotyping neutral and drug resistance-associated markers. Antimicrob Agents Chemother. 2011 Sep;55(9):4479–81. Unselected pregnant women of any age, gestational age, and parity, attending the ANC clinic at each study site, independently of parasitological or disease status, were invited to participate, and after signing a written informed consent were recruited into the study. There were four study visits: recruitment visit, coinciding with an ANC visit; two subsequent scheduled ANC visits one month apart; and at delivery. Project personnel were trained on study procedures that were standardized across all study sites. At each study visit and regardless of the presence of symptoms suggestive of malaria, a capillary blood sample was collected to prepare two thick and thin blood smears and two filter papers (Whatmann 3MM) to determine Plasmodium parasitemia. From the same samples the hemoglobin (Hb) concentration was measured at enrolment, at scheduled ANC study visits, at delivery and at any other time the woman was suspected to have malaria. In addition, demographic, obstetric and clinical information were recorded on standardized questionnaires. Pregnant women were encouraged to deliver at the study health facility. In case of home delivery, they were advised to come to the study health facility within the first week of giving birth. A passive surveillance system to identify study women presenting with clinical malaria was set up at each study health facility. If the women reported any signs and/or symptoms suggestive of clinical malaria, a capillary blood sample was collected to prepare two thick and thin blood smears and two filter papers to determine Plasmodium parasitemia, and for determination of Hb concentration. At delivery, the pregnancy outcome and the clinical information on the mother and the neonate were collected. Placental blood from all women delivering at the health facility was collected onto filter paper for PCR molecular analysis, and two impression smears were stained with Giemsa and read following standard procedures for parasitemia determination. [21] For the preparation of the impression smears a 2.5×2.5 cm3 sample from the placenta, that should include the full thickness of the tissue from the maternal to the fetal side, was cut and put in contact with the slide after being dried with a piece of filter paper. In a randomly selected subsample of women (10%) a placental biopsy was collected, following a similar procedure (2.5×2.5 cm3 sample) to that for the preparation of the impression smear. The biopsy was kept at 4°C in 50 mL of 10% neutral buffer formalin, processed for histological examination, and stained with haematoxylin and eosin as previously described. [22] The histological examination of placental biopsies for malaria infection was performed by local pathologists trained for the purposes of the study (S1 File). Cord blood samples were collected after birth for parasitemia determination. As soon as the umbilical cord was clamped, cut, and separated from the newborn study trained staff collected 5mL cord blood sample taken from the cord with a syringe and needle. Cord blood was used for preparation of two thick and thin blood smears to determine Plasmodium parasitemia, and two filter papers to perform PCR molecular analysis. Newborn samples were collected by heel prick after medical assessment was complete and within the first 12 hours of life also for parasitemia determination. In GT, BR and PNG gestational age was assessed by the Ballard’s method for deliveries that occurred at the health facility within the first 72 hours after birth [23]. In CO and IN gestational age was assessed by ultrasound assessment performed at enrolment. All newborns were weighed on a digital scale, accurate to the nearest gram, within the first 2 hours of life and examined for any clinical abnormalities. For deliveries occurred at home, the study personnel obtained information on the pregnancy outcome through home visits. All women and newborns with malaria infection or anemia were treated according to the national guidelines in each country. Strategies for malaria control in pregnancy differed across study sites. While these relied on active detection of infection with microscopy at each ANC visit in some countries, in other countries passive detection of cases, or weekly prophylaxis with chloroquine (CQ) until delivery were carried out (Table 1). At enrolment a subsample of 1500 women (300 per study site) were randomly selected for the determination of the prevalence of P. vivax and P. falciparum infection by PCR methods. Likewise, a subsample of 1500 women (300 per study site) was randomly selected at delivery for the same purpose. The 1500 women from whom samples were obtained at delivery were different from those selected at enrolment. A total of 500 placental samples (100 per site) and 500 cord blood samples (100 per site) were also analysed following same methodology. A simple random sampling method was used for the selection of women in each country, and for each time point and compartment. The sample size for prevalence by PCR was agreed among study investigators according to preliminary results of the first 100 selected samples in each site that showed a prevalence of P. vivax monoinfection of 7.5% by PCR, and on the availability of resources for molecular analyses in the study. PCR assays were not performed on newborn blood samples. To estimate the impact of submicroscopic P. vivax and P. falciparum infections on maternal anemia and low birth weight (LBW), a nested case-control study was conducted. All countries contributed to this pooled analysis. A definition for moderate-severe anemia as Hb less than 8g/dL was agreed for the purpose of the analysis. All moderate-severe anemia cases (n = 342), and LBW cases (<2500g) (n = 327) existing across countries for which a blood smear and a filter paper were available, were included in the case control analysis. A total of 414 controls to anemia cases, and a total of 410 controls to LBW cases, were randomly selected. Logistic regression models used to evaluate the impact of P.vivax submicroscopic infections on anemia and LBW were adjusted by site and P. falciparum infections. Similarly, for the evaluation of the impact of P. falciparum submicrocopic infections, models were adjusted by site and P. vivax infections. Giemsa-stained thick and thin blood slides were read onsite in all countries following WHO standard quality-controlled procedures to establish parasite presence and density of Plasmodium asexual stages. [24] Two independent expert malaria microscopists read all slides and results were registered in two separate forms. Discrepant results (positive vs. negative) were resolved by a third reading done by a different microscopist. A blood slide was declared as negative only when no parasites were found after reading 200 fields. Results were expressed in parasites/μL after counting the number of parasites per 500 white blood cells or reaching 500 parasites; counting was normalized using estimated leukocyte counts of 8000/μL. External validation of a blood slides subsample (100 slides per country) was done at the Hospital Clinic and at the Hospital Sant Joan de Deu, in Barcelona, Spain. Hb was measured by Coulter Counter (except in PNG where it was done by Hemocue, HemoCue, Ltd, Angelhom, Sweden; accuracy of 0.1 g/dL) using 50–100 μL collected in a 0.5mL EDTA tube (microtainer). Molecular detection of Plasmodium species in samples from CO, GT, BR and PNG was performed by Real Time PCR at the Istituto Superiore di Sanità (ISS) in Rome, Italy. Samples from India were analysed, due to local regulatory requirements, at the International Center for Genetic Engineering and Biotechnology (ICGEB), in New Delhi using the same protocol as that at ISS but adapted for the sake of instrument sensitivity (3rd step at amplification was 72°C for 25 sec instead of 72°C for 5 sec). DNA was extracted from whole blood-spot filter paper from maternal peripheral blood collected at ANC, at delivery and during passive case detection, and from placental and cord blood using Purelink Genomic DNA Kit (Invitrogen). P. vivax and P. falciparum infections were detected with a LightCycler 480 system (Roche). Species-specific primers and Taqman probes were selected from the sequence of the small 18S rRNA subunit as previously described by Veron et al 2009. [25] Briefly, pre-incubation was at 95°C for 10 min; amplification at 95°C for 10 sec, 50°C for 20 sec and 72°C for 5 sec for 50 cycles. All reactions were in duplicate in a final volume of 20 μL. The sensitivity of the PCR assay performed at the ISS for detection of P. falciparum infections was between 10–100 times higher in comparison to the microscopy. An external validation of the PCR methods used by the ISS and the ICGEB, for detection of P. vivax and P. falciparum species, was performed by the Malaria in Pregnancy Consortium (MiPc) (http://www.mip-consortium.org) in a subset of 20 samples. Internal validation between ISS and ICGEB was also done. A standardised system for data entry, data management, and statistical analysis was established. All clinical and laboratory data were collected using standardised questionnaires. The data collection and management was performed using the OpenClinica open source software, version 2.0. Copyright OpenClinica LLC and collaborators, Waltham, MA, USA, www.OpenClinica.com. All data were double entered by two independent data clerks at each of the sites. There was a specific URL link to access the data entry software. In PNG data were doubled-entered into form-specific databases (FoxPro 9·0, Microsoft, USA). Validation and cleaning were done using the same software, and statistical analysis was performed using Stata 13 (Stata Corporation, College Station, TX, USA). Differences between proportions were compared using the Pearson′s chi-squared test or Fisher's exact test depending on type of variables. For continuous variables, Student’s T-tests were used to compare the groups. Incidence rates were calculated as the number of new episodes/person-year at risk using the Poisson distribution in the exposed and unexposed groups, with primigravid women as the comparator group. The impact of P. vivax infection on maternal and newborn health was determined through a multicenter-pooled analysis. The case control study for submicroscopic infections with and anemia was analyzed using logistic regression models. We adjusted all regression models for possible operational confounding variables such as country and previous malaria episodes. In the analyses of P. falciparum infections, they were included in the models as being free of P. vivax (see definitions section). Likewise, in the analyses of P. vivax infections, they were included as being free of P. falciparum. Multivariate analyses were performed by a forward-stepwise procedure, using p0.10 from the likelihood ratio test, as enter and remove criteria respectively. Results from the estimated models were expressed as OR and 95% CI. Missing values were coded as such and excluded from analysis. P. vivax microscopic monoinfection was defined as the presence of asexual P. vivax parasites of any density and absence of other Plasmodium species on the blood smear. P. vivax clinical malaria episode was defined as the latter plus any signs or symptoms suggestive of malaria (axillary temperature ≥37.5°C or history of fever in the last 24 hours, headache, arthromyalgias, and/or pallor). P. vivax submicroscopic infection was defined as a PCR that was positive for P. vivax and negative for P. falciparum, with a concomitant blood film negative by microscopy. P. falciparum microscopic monoinfection was defined as the presence of asexual P. falciparum of any density and absence of other Plasmodium species on the blood smear. P. falciparum clinical malaria episode was defined as the latter plus any signs or symptoms suggestive of malaria. P. falciparum submicroscopic infection was defined as a PCR that was positive for P. falciparum and negative for P. vivax, with a concomitant blood film negative by microscopy. The duration of any single malaria episode was estimated as 28 days. Congenital malaria was defined as presence of asexual Plasmodium parasites of any species in the cord blood or in the newborn′s peripheral blood at delivery, regardless of clinical symptoms or signs in the neonate. Placental infection was classified according to a previously established definition. [21] Briefly, acute infection was defined as the presence of parasites, with absent or minimal pigment deposition within fibrin or cells within fibrin, chronic infection as the presence of parasites and a significant amount of pigment deposition, and past infection as the presence of pigment with absence of parasites. Prematurity was defined as gestational age < 37 weeks. LBW was defined as birth weight <2500g.